MS(Online)-2014(12)--Maximizing Stochastic Monotone Submodular Functions
Strong Lefschetz elements of the coinvariant rings of finite Coxeter groups
Keywords: Coxeter group, Weyl group, coinvariant ring, strong Lefschetz property, flag variety, hard Lefschetz theorem.
Theorem 1. Let W be a finite Coxeter group which does not contain irreducible components of type H4. Then a homogeneous element ℓ of degree one is a strong Lefschetz element if and only if ℓ is not fixed by any reflections of W.
In view of such a property of the cohomology ring, the strong Lefschetz prop-
erty is defined as follows:
Definition 3. A graded ring R =
m d=0
Rd
havபைடு நூலகம்ng
a
symmetric
tion 7. see [GH78], e.g.) tells us that the multiplication by the class of the K¨ahler form induces an isomorphism between Hi(X, R) and H2 dimC X−i(X, R).
成人高考升专科英语试卷
一、语音知识(共5小题,每题1.5分,共7.5分)1. 请选出下列单词中划线部分发音与其他三项不同的单词。
A. farmB. fanC. farD. family2. 请选出下列句子中划线部分发音与其他三项不同的句子。
A. She is a teacher.B. I have two brothers.C. He works hard.D. They are students.3. 请选出下列单词中划线部分发音与其他三项不同的单词。
A. mapB. catC. mapleD. match4. 请选出下列句子中划线部分发音与其他三项不同的句子。
A. It's a beautiful day.B. I like apples.C. She is in the classroom.D. They are watching TV.5. 请选出下列单词中划线部分发音与其他三项不同的单词。
A. homeB. hopeC. holeD. hold二、词汇与语法知识(共15小题,每题1.5分,共22.5分)1. Fill in the blanks with the proper form of the given verb in the brackets.They _______ (be) at the library when I called them.2. Choose the correct form of the given word to complete the sentence.She _______ (be) a teacher for 10 years.3. Choose the correct word to complete the sentence.He _______ (go) to the movies last night.4. Choose the correct word to complete the sentence.She _______ (not go) to the party because she was ill.5. Choose the correct word to complete the sentence.The weather _______ (not be) good today.6. Choose the correct word to complete the sentence.I _______ (not go) to the park because it's raining.7. Choose the correct word to complete the sentence.He _______ (not work) hard enough to pass the exam.8. Choose the correct word to complete the sentence.They _______ (not go) to the concert last night.9. Choose the correct word to complete the sentence.The movie _______ (not be) interesting.10. Choose the correct word to complete the sentence.She _______ (not go) to the party because she was tired.11. Choose the correct word to complete the sentence.The weather _______ (not be) good yesterday.12. Choose the correct word to complete the sentence.He _______ (not work) hard enough to pass the exam.13. Choose the correct word to complete the sentence.They _______ (not go) to the concert last night.14. Choose the correct word to complete the sentence.The movie _______ (not be) interesting.15. Choose the correct word to complete the sentence.She _______ (not go) to the party because she was tired.三、完形填空(共15小题,每题2分,共30分)Read the following passage and choose the best answer for each blank.When I was a child, I lived in a small town. There was a river running through the town, and it was my favorite place to play. Every day, I would go to the river with my friends, and we would fish, swim, and play games.One day, my father told me that the river was polluted, and it was not safe for us to play there anymore. I was sad and worried, because I loved the river so much. I decided to do something about it.I started by talking to my friends and asking them to join me in cleaning the river. We collected garbage and plastic waste from the riverbank and threw it in the trash. We also wrote letters to the local government, asking them to take action to clean the river.After a few months, the government finally responded. They began to clean the river and restore its natural beauty. The river became clean again, and we could play there without worrying about pollution.I learned a valuable lesson from this experience. I realized that even a small group of people can make a difference if we work together. I am grateful for the opportunity to help clean the river and make it a safe place for everyone to enjoy.1. The author's favorite place to play was the ___________.A. parkB. beachC. riverD. forest2. The author's father told him that the river was ___________.A. cleanB. pollutedC. dirtyD. beautiful3. The author decided to do something about the pollution by ___________.A. ignoring itB. writing letters to the governmentC. cleaning theriver himself D. moving to a different town4. The author and his friends collected garbage and ___________.A. recycled itB. threw it in the trashC. burned itD. buried it5. The author and his friends wrote letters to the ___________.A. local governmentB. school principalC. mayorD. river6. The government began to clean the river and ___________.A. ignored the problemB. restored its natural beautyC. destroyed the riverD. left it alone7. The river became clean again, and the author could play there___________.A. without worrying about pollutionB. with his friends onlyC. without his fatherD. with the government8. The author learned a valuable lesson from this experience, which was ___________.A. the importance of working togetherB. the importance of being aloneC. the importance of ignoring problemsD. the importance of moving to a different town9. The author is grateful for the opportunity to help clean the riverand ___________.A. move to a different townB. write letters to the governmentC. make the river a safe place for everyoneD. ignore the problem10. The author's story shows that ___________.A. a small group of people can't make a differenceB. working together can solve problemsC. it's important to move to a different townD. the government should always take action11. The author's friends ___________.A. helped him clean the riverB. ignored the problemC. moved to a different townD. wrote letters to the government12. The author's father ___________.A. supported his decision to clean the riverB. didn't believe the river was pollutedC. was worried about the river's pollutionD. ignored the problem13. The author's story took place in ___________.A. a big cityB. a small townC. a forestD. a beach14. The author's friends ___________.A. collected garbage and wrote letters to the governmentB. ignored the problemC. moved to a different townD. wrote letters to the mayor15. The author's story shows that ___________.A. a small group of people can't make a differenceB. working together can solve problemsC. it's important to move to a different townD. the government should always take action四、阅读理解(共15小题,每题3分,共45分)阅读下列短文,然后回答问题。
时间序列 数据清洗和预处理 数据分解 box-cox方法
时间序列数据清洗和预处理数据分解box-cox方法1. 引言1.1 概述:时间序列数据分析是一种广泛应用于各个领域的数据分析方法,它能够揭示时间相关性和趋势,帮助我们预测未来趋势、进行决策和制定策略。
然而,时间序列数据经常存在一些问题,如噪音干扰、缺失值以及非线性等,这些问题会对分析结果的准确性产生负面影响。
因此,在进行时间序列数据分析之前,我们需要进行数据清洗和预处理的工作。
本文将重点讨论时间序列数据清洗和预处理的方法。
1.2 文章结构:本文共分为五个主要部分。
首先,引言部分介绍了文章的概述、目的和重要性。
第二部分将详细介绍时间序列数据清洗和预处理的过程,包括数据收集和获取、数据清理和缺失值处理以及数据平滑和去噪。
第三部分将介绍常用的时间序列数据分解方法,包括经典分解方法和基于小波的分解方法。
第四部分则着重探讨Box-Cox转换方法在时间序列数据预处理中的应用,并提供实现方法和应用案例分析。
最后,在结论与展望部分对本文进行总结并提出改进方向展望。
1.3 目的:本文的目的是探讨时间序列数据清洗和预处理的方法,以及容易忽视但重要的Box-Cox转换方法在时间序列数据分析中的应用。
通过深入了解和研究这些方法,读者将能够更好地理解如何有效地处理时间序列数据,降低噪音干扰、处理缺失值,并提高对数据趋势和相关性的理解能力。
此外,我们还将通过实际案例分析来展示这些方法在实际问题中的应用效果,帮助读者更好地理解其实际价值和应用场景。
最终,我们期望本文对时间序列数据分析领域的从业人员和学术研究者有所帮助,并为进一步研究和应用提供指导。
2. 时间序列数据清洗和预处理2.1 数据收集和获取数据收集是时间序列分析中的第一步,它涉及到获取可用于分析的原始时间序列数据。
常见的数据收集方法包括实时采集、历史数据提取和数据库查询等。
在进行数据收集之前,需要明确所需的时间范围、采样频率以及目标变量等。
2.2 数据清理和缺失值处理在时间序列数据中,经常会遇到许多问题,如异常值(outliers)、噪声(noise)以及缺失值(missing values)等。
微软消息分析器:一款高级网络包分析器说明书
Neil B MartinT est Manager WSSC-Interop and T oolsMicrosoft CorporationMicrosoft Message Analyzer Packet Analysis at a Higher LevelContent•Packet Analyzer -review •Abstracting views of protocols •Alternative data sources •ETW•Remote Capture•Bluetooth•USB•Evtx•Logs filesMessage Analyzer –What is it?• A packet analyzer is a computer program or a piece of computer hardware that can intercept and log traffic passing over all or part of a network•Packet analyzers capture network packets in real time and display them in human-readable format•WireShark, Microsoft NetMon3.4•These tools are dissectors•If they recognize a packet they dissect it and display the inner fields of the packet•The parsers are written based on the protocol specifications or in some cases through reverse engineering of the protocols whenno specification is available•Dissection and Abstraction•We want to allow a higher level of abstraction view of protcols•Pattern Matching•Match up request/response pairs where possible•Called an operation•Different Viewers and Charts•Addressing many of the challenges of diagnosing modern networks •Protocol Validation•Identify packets that do not match the specification•Data capture from multiple sources•NDIS, Bluetooth, USB, Windows Firewall Layer, Web Proxy•Header only network capture•Reduce data in volume scenarios•Correlation of data across multiple data sources and logs•Load and display multiple data source•Message Analyzer captures ETW •ETW -Event Trace for Windows •Message Capture from:•Traditional NDIS traffic from the Network Adapter •Windows Filtering Platform 9aka Firewall)•Web proxy•USB ports•Bluetooth•Windows SMB Client•Windows SMB Server ……•E vent T racing for W indows ETW•High-resolution (<<100µs)logging infrastructure allows any component to tell the outside world what it is currently doing by firing ETW events.• A powerful diagnostic tool to log every methods/lines inside the code with reasonable performance fordebugging/troubleshooting.•MSDN on ETW/en-us/library/bb968803(VS.85).aspxAll Windows ETW Sources are available to Message Analyzer•Capability to perform remote capture •Select machine and give credentials•Collect data via ETW from NIC on remote machine•Powerful, extensible viewing and analysis •Browse, Select, View•Browse for messages from various sources (live, or stored)•Select a set of messages from those sources by characteristic(s)•View messages in a provided viewer, configure or build your own • A new high-level grid view•High level “Operations” view with automatic re-assembly•“Bubbling up” of errors in the stack to the top level•Ability to drill down the stack to underlying messages and/or packets•On the fly grouping, filtering, finding, or sorting by any message property •Payload rendering•V alidation of message structures, behavior, and architecture•Does the protocol comply with the specifications?•Over 450 published specifications for Windows Protocols(as of Windows 8.1)(/en-us/library/gg685446.aspx)Available online and as PDFContinue to publish new documents with each release of Windows •Continue to develop tools and technology to aid with the development of protocol documents, parsers and test technologyHow to get MA: /en-us/download/details.aspx?id=40308 How to get help: Blog,Operating Guide, T echnet Forum for Message Analyzer •We invite you to Explore Message Analyzer•Connect Community•https:///site216/。
REPSOL_VETTING_PROCESS_AND_CRITERIA_2014_tcm11-689923
III. Effective Date
01 Aug 2014
Page 2 of 31
REPSOL VETTING PROCESS AND CRITERIA
IV. DEFINITIONS For the purpose of these procedures, the following definitions apply: Acceptable means the vessel can be used within the scope described above, and is the only rating that allows such use. This rating results from a favourable assessment based on information that we have deemed positive and sufficient. The rating of the vessel may be affected by relevant modifications concerning safety and operational systems, changes of name, technical operator, crew, flag, etc., as well as any incident, casualty or terminal negative feedback report, PSC detention or Memoranda or condition of Class. (See also “Vetting Assessment”.) Barge, for the purpose of these procedures, means a vessel carrying goods in rivers, inland navigation, lakes and ports, not sailing on open sea or bays and restricted by Flag Administration to inland water navigations. EBIS Barge: for the purpose of these procedures, means a vessel carrying goods in European rivers, not sailing on open sea or bay CAP (Condition Assessment Programme).- Independent and thorough scheme of inspections of the actual condition of a vessel. It is applicable as established in the present Vetting Process and Criteria and as defined in the Rules of the Classification Societies members of IACS. Cargo means any kind of material subject to a contract of transportation, mainly crude oil, oil products, chemical products, LPG, LNG, Lubricants, Liquid fertilizers and dry bulk cargoes. Charter Party means contract of affreightment signed between shipowner and charterer when hiring a vessel for the carriage of goods. Chief Officer and 2nd. Engineer terminology considered equivalent to 1st. Officer and 1st. Asst. Engineer for the purpose of these procedures. COA vessel means vessels included in a contract of affreightment to lift a fixed or determinable quantity of cargo of a specified type over a given period of time. EBIS the European Barge Inspection Scheme, is used to evaluate barges, tugs and dumb barges used to distribute oil and chemicals within Europe ESP (Enhanced Survey Programme).- It is applicable as established in SOLAS XI1/2 and as defined in Resolution A.744 (18). Heavy grade Oil: o o o crude oils, having a density at 15º C higher than 900 kg/m3; oils, other than crude oils, having either a density at 15º C higher than 900 kg/m3 or a kinematic viscosity at 50 º C higher than 180 mm2/s; or; bitumen, tar and their emulsions.
CDA_LEVEL_1试题及答案
CDA LEVELⅠ业务分析师_模拟题:一、单选1.北京市统计局发布2014年度全市职工平均工资为77560元,月平均工资为6463元.众多网友直呼“被平均”,你认为下面哪种统计量测度平均工资会更被信服()A几何平均数B众数C极差D中位数答案:D2.某企业2000年实现利润为200万元,2005年为300万元,则年平均增长速度为()A.5%B.11%C.10%D.8.4%答案:D3.当一组数据属于左偏分布时,则()A.平均数、中位数与众数是合而为一的B.众数在左边、平均数在右边C.众数的数值较小、平均数的数值较大D.众数在右边、平均数在左边答案:D4.作为一家制造类企业,以下哪个图适合比较不同产品各年的销售变化情况()A.分组饼形图B.堆叠面积图C.堆叠柱形图D.分组柱形图答案:D5.横截面数据主要注意以下哪个数据问题()A.异方差B.不独立C.非正态分布D.不随机答案:A.6.以下叙述正确的是()A.极差较少受异常值的影响B.四分位差较少受异常值的影响C.方差较少受异常值的影响D.标准差较少受异常值的影响答案:B。
7.某汽车品牌预测到未来不同型号汽车的利润率和销售量会发生变化,希望根据利润最大化得到最优产量,这属于哪类数据分析过程()A.预测型建模B.预报C.优化D.报警答案:C8.为研究某种减肥茶减肥效果是否显著,可以采用()分析方法。
A、单样本t检验B、两独立样本t检验C、两配对样本t检验D、方差分析答案:C9.中心极限定理的假设不包括()A.样本相互独立B.样本具有相同分布C.样本足够大D.样本服从正态分布答案:D10.下列场合下,()适合用t检验统计量A.样本为小样本,且总体方差已知B.样本为大样本,且总体方差已知C.样本为小样本,且总体方差未知D.样本为大样本,且总体方差未知答案:C11.方差分析表表如下,值20应填在哪个位置上()方差来源离差平方和自由度均方差F值组间14245.8334748.61 2.16组内A B C总和D23答案:B12.某信用卡公司为了分析客户教育程度对授信额度是否有显著性差异,已知教育程度分为5种,每个教育程度取30个客户,则因素的水平为()A5B6C30D15013.给出下列结论:(1)在回归分析中,可用指数系数R方的值判断模型的拟合效果,R方越大,模型的拟合效果越好;(2)在回归分析中,可用残差平方和判断模型的拟合效果,残差平方和越大,模型的拟合效果越好;(3)在回归分析中,可用相关系数r的值判断模型的拟合效果,r越小,模型的拟合效果越好;(4)在回归分析中,可用残差图判断模型的拟合效果,残差点比较均匀地落在水平的带状区域中,说明这样的模型比较合适.带状区域的宽度越窄,说明模型的拟合精度越高.以上结论中,正确的有(B)个.A.1B.2C.3D.414.下列关系中,属于正相关关系的有()A.合理限度内,施肥量和平均单位产量之间的关系B.产品产量与单位产品成本之间的关系C.商品的流通费用与销售利润之间的关系D.流通费用率与商品销售量之间的关系答案:A15.下列关于SQL的说法错误的是()A SQL对大小写不敏感B SQL为非过程化编程语言C不同的数据库的SQL完全一致D一种数据库查询和程序设计语言,用于存取数据以及查询、更新和管理关系数据库系统答案:C16.要查询book表中所有书名中以“中国”开头的书籍的价格,可用()语句。
企业道德和社会责任 Business Ethics and Social Responsibilit
What Business Areas Does CSR Cover
• Ethics - discretionary actions. • Moral righteousness.
• Strategic brand management - Brand building, Brand insurance.’
Corporate social responsibility (CSR) is concerned with the ways in which an organisation exceeds its minimum obligations to stakeholders specified through regulation
• This poses a challenge as there are many different stakeholders with different, perhaps conflicting, expectations.
• Managers will need to take a view on: i. Which stakeholders have the greatest
development in LDCs. • Importance of global brands and corporate
reputations.
11
Key Drivers in CSR
• Changing social expectations - Consumers and society in general expect more from the companies whose products they buy. This sense has increased in the light of recent corporate scandals, which reduced public trust of corporations, and reduced public confidence in the ability of regulatory bodies and organisations to control corporate excess.
人教版英语七年级上册Unit1 SectionB1(1a-1d)课件
Parrot Beijing roast duck the UK Singapore guitar tennis
parro t
1a Wrhiatet dthoeywouorldeasrin athbeoubtoxthuentdweor stthuedceonrtrse?ct pictures.
Parrot Beijing roast duck the UK Singapore guitar tennis
in the UK. Now, I live in
Beibjinigg with my parents. My
fdauvcopkua.rIrittr'esofgtoroedati!s IBleikijeinmg ursoiacst
and I play the guitar in the
school band. Would you like
The Lion City is Singapore’s nickname(昵称). Merlion Park is the symbol of Singapore.
鱼尾狮 公园
the UK
Big Ben is the symbol of Singapore
the national flag of the UK
country
hobby
country
I think he/he lives in__S_in_g_a_p__o_re__. She has a __p_a_r_r_o_t___. She likes _te_sn_pn_o_is_rt.
pet
I think he / she likes playing the __g_u_i_ta_r_. He is from
Msaastricht共识解读
治疗方案选择
新共识修改的重点 H.pylori耐药率上升,既往作为一线方案的标准三联
疗法的根除率已低于或远低于80%,而H.pylori感染 作为感染性疾病,根除率应高于90%。
13
影响推荐方案的因素
主要因素有:分子试验法检测Hp耐药率高低、铋剂(不 少国家无铋剂)和呋喃唑酮(发达国家无该药)的可获得 性:利福布丁是否用于根除Hp(WG0和亚太共识推 荐)。
随着标准三联方案根除率的下降,近年陆续出现了一些 新的方案,包括序贯疗法(sequentialtherapy)、 伴同疗法(concomitant therapv)和左氧氟沙星三 联方案(1evofloxacin triple therapv)。共有5种 根除方案。
15
Maastricht-Ⅳ共识推荐这5种方案
16
5种根除方案
① 标准三联方案: 有2个方案:
✓ 质子泵抑制剂(PPI)+克拉霉素+阿莫西林 ✓ PPI+克拉霉素+甲硝唑
由于方案中均包含克拉霉素,故在克拉霉素耐药率高地 区的应用受到限制。我国报道的克拉霉素耐药率为 27%~38%。
17
5种根除方案
②铋剂四联方案: ✓ 经典铋剂四联方案由铋剂+PPI+四环素+甲硝唑组成。
态,但根除Hp后的胃酸分泌变化并未显示有临床意义。
4
Hp与胃食管反流病 ✓ Hp状态对胃食管反流病(GERD)症状严重性、是否
复发及疗效无影响。根除Hp不会加重原本已存在的 GERD,也不会影响疗效。 ✓ 流行病学研究表明,Hp感染率与 GERD 严重性和食 道腺癌发病率负相关。
5
Hp与肠化生 根除Hp有可能预防胃癌。越来越多的证据表明,根除 Hp后的胃体功能可获改善,但尚不能确定这是否因萎 缩性胃炎逆转所致。尚无证据表明根除Hp能逆转肠化 生。在慢性胃炎向胃癌进展的过程中,可能存在所谓的 “不可逆转点”,对于已经超过该点的患者,根除Hp 虽可延缓肠化生,但并不能彻底预防胃癌。
chapter-03
Chapter 3. Specialized TerminologiesThe specialized vocabulary used in various scientific disciplines has precise meaning to those engaged in that discipline, but occasionally a different meaning to scientists practic-ing a different discipline.Professional societies try to present information in their journals as clearly as possible to their readers. This manual should be used as a primary source for conventions and style in all ASA, CSSA, SSSA publications. Other style manuals supplement this manual, including Scientific Style and Format (CSE, 2006), the ACS Style Guide (Coghill and Garson, 2006), the Chicago Manual of Style (UCP, 2010), and the US Government Printing Office Style Manual, 2008 (USGPO, 2008). Authors are also encouraged to study recent issues of ASA, CSSA, SSSA journals and books for the general style and format used.Except as new terminology itself forms the content of a paper (as in reports on gene names for a given crop, or proposals for new evaluation scales), authors should avoid making up new terms. If new developments seem to call for new terms, authors should still consult others who normally work in the field in question before trying to devise a new terminology. It is also wise to do a literature search for related materials published by the Societies and elsewhere to see if a consensus on terminology exists or is emerging. In some cases, simply consulting a good dictionary, or the chapters on specialized terms in the major scientific style manuals, is enough to resolve a terminology question.A number of committees of ASA, CSSA, and SSSA have studied terminology in specialized fields and in many cases have indicated a preference.CROP SCIENCE GLOSSARYGlossary of Crop Science Terms is available on the CSSA Website (www.crops. Theorg/publications/crops-glossary).Earlier lists of terms compiled by various committees on crop terminology were pub-lished in Crop Science (Leonard et al., 1968; Shibles, 1976). These reports cite relevant articles and lists published in related fields and include previously published reports issued by earlier committees. In addition, letters in the journal may comment on various aspects of terminology (e.g., Dybing, 1977).SOIL SCIENCE GLOSSARYGlossary of Soil Science Terms is available both in hard copy (SSSA, 2008) and Theon the SSSA Website (/publications/soils-glossary). It contains definitions of more than 1800 terms, a procedural guide for tillage terminology, an outline of the US soil classification system, and the designations for soil horizons and layers. Obsolete terms are noted as such.SPECIALIZED TERMINOLOGYCrop Growth Staging ScalesThe CSSA Ad Hoc Committee on Growth Staging for CSSA Publications (C392.1) in 1996 developed a list of growth staging scales for society publications. The committee recommends that staging scales be used in all ASA, CSSA, SSSA publications when refer-ring to the morphological development stage of plants. References for crop-specific scales recommended by the committee for some major crops are listed in Table 3–1. This list is not intended to include all scales in the literature, but rather the most recent versions for some major crops. If no staging scale exists for a crop, it is recommended that the BBCH (BASF–Bayer–Ciba-Geigy–Hoechst) scale be used (Lancashire et al., 1991).Copyright © ASA–CSSA–SSSA, 5585 Guilford., Madison, WI 53711, USA.Soil IdentificationAll soils discussed in publications of ASA, CSSA, and SSSA should be identified according to the US soil taxonomic system or World Reference Base for Soil Resources the first time each soil is mentioned. Taxonomic identification given in the abstract need not be repeated in the text. If possible, give the series name in addition to the family name. If the series name is not known, give the family name. If the family name is not known, give the subgroup or a higher category name. At a minimum, specify the great group (the one-word name that is the third-highest taxon, beneath suborder and order; e.g., Dystroxerepts, Fragiudalfs, Medisaprists, Natrargids).The descriptive name may be in the singular or plural, according to meaning. Use the singular form if the reference is to a single pedon or polypedon or to a single class.E xamplEs:• The soil material used in this study was collected from the A horizon of a Brookston pedon (a fine-loamy, mixed, mesic Typic Argiaquoll).• A Cisne soil, fine, smectitic, mesic Vertic Albaqualf, was described and sampled at this site.• Criteria for the Typic Hapludult subgroup were examined.• Ontario soils, in the fine-loamy, mixed, mesic Glossoboric Hapludalf family, were studied in greater detail.Use the plural form in reference to several or all of the soils (polypedons) of a class.E xamplEs:• Soils of the Ramona series (fine-loamy, mixed, thermic Typic Haploxeralfs) were treated.• All soils used in the experiments are Typic Dystrochrepts.Table 3–1. Some recommended staging scales and sources for ASA, CSSA, SSSA publications. Recommendations are as developed by the Ad Hoc Committee on Growth Staging for CSSA publications (C392.1) in 1996.Crop CitationAlfalfa Kalu and Fick (1981)Fick and Mueller (1989)†Corn Ritchie et al. (1996)Cool-season forage grasses Haun (1973)Moore et al. (1991)Cotton Elsner et al. (1979)Red clover Ohlsson and Wedin (1989)Small-grain cereals Haun (1973)Zadoks et al. (1974)Tottman (1987)‡Sorghum Vanderlip and Reeves (1972)Soybean Fehr and Caviness (1977)Ritchie et al. (1994)§Stoloniferous grasses West (1990)Sunflower Schneiter and Miller (1981)Warm-season forage grasses Moore et al. (1991)Sanderson (1992)All crops and weeds Lancashire et al. (1991)¶† Enhancement of Kalu and Fick (1981).‡ Enhancement of Zadoks et al. (1974).§ Enhancement of Fehr and Caviness (1977).¶ The BBCH (BASF–Bayer–Ciba-Geigy–Hoechst) scale as presented by Lancashire et al. (1991) can be used for all other crops and weeds.Copyright © ASA–CSSA–SSSA, 5585 Guilford Rd., Madison, WI 53711, USA.For field experiments, the soil present in the plots or fields should be identified, preferably as phases of soil series so that surface texture and slope are known in addition to profile properties. Any dissimilar inclusions that are present also should be named and their extent suggested. It also may be appropriate to name and briefly describe the common soils of the area surrounding the study site. Use the present tense if the soil still exists or reasonably is thought to still exist. E xamplE:The 5-ha study area is mapped as Yolo silt loam, 0 to 2% slopes. The Yolo soils are fine-silty, mixed, nonacid, thermic Typic Xerorthents. Small areas of Cortina very gravelly sandy loam soils (loamy-skeletal, mixed, superactive, nonacid, thermic Typic Xerofluvents) occupy about 10% of the study area.The US taxonomic system should be identified as the US soil taxonomy at first use, after which it may be referred to as Soil Taxonomy. Amendments to Soil Taxonomy (Soil Survey Staff, 1999) have been issued in the National Soil Survey Handbook (http://www. /wps/portal/nrcs/detail/soils/survey/?cid=nrcs142p2_054242) and in Keys to Soil Taxonomy (Soil Survey Staff, 2014). Additional issues of the handbook and new ver-sions of the keys manual can be expected. Updated versions of these and other resources are available online at the Soil Survey home page ().If possible, consult with members of the National Cooperative Soil Survey (NCSS) and check the current USDA-NRCS official soil series descriptions (https:///osdname.aspx) for proper identification of soil designations and nomenclature for soil horizons.For soils outside the United States, authors are encouraged to give soil identifica-tion according to Soil Taxonomy in addition to the identification in their national system.E xamplE:Soil at the site is a Hythe clay loam, classified as a fine, montmorillonitic, frigid Mollic Cryoboralf in the USDA classification (Soil Survey Staff, 1994) and a Gray Luvisol in the Canadian classification (Canada Soil Survey Committee, 1978).Munsell Color NotationMunsell color notations may be used alone in text, tables, or figures. First mention in the abstract or text may be accompanied by the appropriate word descriptions in paren-theses, thus: 10YR 5/4 (yellowish brown).Light Measurements and PhotosynthesisPublications of the ASA, CSSA, and SSSA use the radiometric system with SI units denoting the energy or the quantum content of the radiation used by plants. (See also Chapter 7.)Terms recommended by the Committee on Crop Terminology for the expression of photosynthetic energy and photosynthetic capacity are as defined by Shibles (1976). These terms, with their suggested abbreviations and units, are as follows.1• Photosynthetically active radiation (PAR): radiation in the 400- to 700-nm wave-band.1 Since 1976, the Societies have abandoned the einstein (a name for 1 mole of photons) in favor of the mole. Note that in the original Shibles (1976) article, the typographic errors “nE” and “nmol” are to be read as µE and µmol.Copyright © ASA–CSSA–SSSA, 5585 Guilford., Madison, WI 53711, USA.• Photosynthetic photon flux density (PPFD): the number of photons in the 400- to 700-nm waveband incident per unit time on a unit surface. Suggested units: µmol m−2 s−1.• Photosynthetic irradiance (PI): the radiant energy in the 400- to 700-nm waveband incident per unit time on a unit surface. Suggested units: W-m−2.• Apparent photosynthesis (AP): photosynthesis estimated indirectly and uncor-rected for respiratory activity. The term apparent photosynthesis is preferred to ‘net photosynthesis’ or ‘net assimilation’, because the latter terms imply measurement of a photosynthetic product.• CO2 exchange rate (CER): The net rate of carbon dioxide diffusion from (−) or to (+) an entity, such as a plant tissue, organ or canopy, a soil surface, etc. Suggested units: µmol cm−2 s−1. (Use this term instead of "net CO2 exchange" except in the rare instance when the measurement does not involve a rate.)Reporting PAR in photon units (PPFD) is preferred to energy units (PI), but both are acceptable. Because irradiance is specifically defined in energy units (W), the term cannot be applied to photon flux density.Abandoned as a term is light intensity to denote the amount of light incident on a surface (Dybing, 1977). The Crop Science editorial board has discontinued the use of the photometric system and units scaled to the response of the human eye.SPECIALIZED TERMINOLOGY IN RELATED FIELDSBiologyIdentify all organisms at first mention. For plants, pathogens, and insects and related pests, give both a common name and the scientific name. For plants, include the authority.E xamplE:Sorghum [Sorghum bicolor (L.) Moench] was. . . .The scientific name, also known as the Latin name, is the two-part genus–species bino-mial—or, for subspecies and varieties, the trinomial. For abbreviations of authorities, the primary source is Authors of Plant Names by Brummitt and Powell (1992). If the first mention is in the abstract, the scientific name need not be repeated in the text. Common names, if they exist and are not in dispute, are used in titles of articles, chapters, and books.For the names of crops, use the singular. Although the ordinary English preference is for terms such as oats, beans, and peas, the formal name of a crop defined by a single genus or species is given in the singular: oat, bean, pea, soybean, and so forth. This rule applies even when discussing multiple types of a crop.For common names that are taxonomically inaccurate, join the parts into a single word. For example, writing "pigeonpea" and "chickpea" as one word indicates that these are not Pisum species; similarly, the absence of a space in the common name indicates that Douglasfir is not an Abies species. The USDA-ARS style (set solid) is preferred to the USDA Forest Service style (hyphenated, the traditional usage for forestry).Correct scientific names are in accordance with published rules. For plants, the International Code of Botanical Nomenclature (McNeill et al., 2006; http://ibot.sav. sk/icbn/main.htm) governs; updates appear in Regnum Vegetabile as mandated by the International Botanical Congress, which meets every six years. For cultivated plants, the rules of nomenclature are published as the International Code of Nomenclature for Cultivated Plants (Brickell et al., 2009; /chronica/pdf/sh_10.pdf).Copyright © ASA–CSSA–SSSA, 5585 Guilford Rd., Madison, WI 53711, USA.A practical guide to these codes and to the standards for animals, bacteria, and viruses is published in Scientific Style and Format (CSE, 2006, Chapters 21–24).The scientific names for larger animals (e.g., sheep) do not need to be given unless germane to the article and/or there may be confusion as to what animal is being discussed. Virus species do not have Latin names, but the name of the virus (as approved by the International Committee on Taxonomy of Viruses) should be written in italics, with the first word capitalized (e.g., Tomato spotted wilt virus).To find up-to-date scientific names, consult one of the major online databases: • https:///gringlobal/taxon/taxonomysimple.aspx for plants, especially economic plants (USDA National Plant Germplasm System, Germplasm Resources Information Network [GRIN] database)• for plants, especially noncrop plants (USDA-NRCS) • /fungaldatabases/index.cfm for fungi (USDA Systematic Botany and Mycology Laboratory; Farr and Rossman, 2012)• /publications/commonnames/Pages/default.aspx for plant disease names (American Phytopathological Society)• h ttp:/// for insect scientific names (Texas A&M University) • /index.asp (International Committee on Taxonomy of Viruses) The International Plant Names Index, a product of a collaboration between the Royal Gardens, Kew, the Harvard University Herbaria, and Australian National Herbarium, is available online (/index.html). (This replaces the Kew Index.) Standard printed reference works for nomenclature include Hortus III (Bailey, 1976) and World Economic Plants: A Standard Reference (Wiersema and León, 1999) for plants; Farr et al. (1989) for fungi; Bergey’s manual (Garrity et al., 2001–2011) for bacteria; and, for viruses, Büchen-Osmond (2003).The terms cultivar and variety are synonymous as applied to names of cultivated plants, but cultivar is strongly preferred, to avoid confusing cultivated variety (a term of convenience) with botanical variety (a subtaxon to species).Crop cultivars must be identified as such at first mention in abstract or text. This identification may be given in one of the following two ways:1. By single quotation marks inside punctuation. E xamplE: ‘Vernal’ al f alfa orMedicago sativa L. ‘Vernal’.2. By use of the word cultivar. E xamplE: the cultivar Vernal.Journal of Plant Registrations publishes articles on registered cultivars, germplasms, parental lines, genetic stocks, and mapping populations. Information on these registrations is also available from the GRIN database (https:///gringlobal/search. aspx), usually with some additional narrative. The database entries include pending regis-trations and are linked to plant variety protection status.Citing Genetic MaterialAuthors of CSSA publications must cite plant introductions, as well as registered cul-tivars, germplasm, parental lines, and genetic stocks when they are mentioned in the text of the Introduction, Discussion, or Characteristics section of research papers. Such genetic materials must also be cited when they are used to develop unreleased genetic populationsCopyright © ASA–CSSA–SSSA, 5585 Guilford., Madison, WI 53711, USA.that are the focus of the research paper, unless the development of the population can be cited more directly. Authors are encouraged to cite the Journal of Plant Registrations if possible. Other sources for citation information include GRIN, maintained by the USDA. Registrations published in Crop Science and the Journal of Plant Registrations are indexed on the GRIN Website at https:///gringlobal/query/query.aspx. A gen-eral search in GRIN is available at https:///gringlobal/search.aspx. Reference ExamplesLewis, J.M., L. Siler, E. Souza, P.K.W. Ng, Y. Dong, G. Brown-Guedira, G.-L. Jiang, and R.W. Ward. 2010. Registration of ‘Ambassador’ wheat. J. Plant Reg. 4:195–204. USDA-ARS National Genetic Resources Program. 1993. Germplasm Resources Information Network (GRIN) database. Festuca arundinacea Schreb. POACEAE ‘Maximize’. National Germplasm Resources Laboratory, Beltsville, MD. http:// /cgi-bin/npgs/acc/display.pl?1444051 (accessed 1 Feb. 2012).Genetics and Molecular and Cell BiologyGenes are named according to established conventions, which vary in part among crops. As an example, a standard for cotton is Kohel (1973). Many of these are summa-rized in Scientific Style and Format (CSE, 2006, p. 298–312); see also the entries for gene and genotype in the New Oxford Dictionary for Scientific Writers and Editors (Martin, 2009). Check with an expert in your field to find the appropriate published standards, including updates. Accepted names of genes are set in italics and may be modified with letters or numbers (with or without superscripts, with or without italics). Proposed names follow the conventions for the crop in question but are set in roman type.Use italics for the variables in ploidy formulas (e.g., 2n = 2x = 42).Spell out amino acids in text, without capitalization. In formulas and sequences, use the abbreviations shown in Table 3–2.For enzymes, follow nomenclature for name and number (Webb, 1992; http://www. /iubmb/enzyme/).As for genetics, the CSE manual (CSE, 2006) is an excellent guide to style for spe-cialized terms and usages in molecular and cell biology, as is the New Oxford Dictionary for Scientific Writers and Editors (Martin, 2009). The Oxford book gives, for example, complete rules for names of restriction enzymes: three letters in italics to identify the source bacterium (e.g., Hin for Haemophilus influenzae, or Bam for Bacillus amylolique-faciens), then letters in roman type to indicate the strain (e.g., d or H), then capital roman numerals to indicate the type of enzyme (e.g., I, II, or III), all leading to characteristic names such as Hin dIII (for enzyme III from strain d of H. influenzae) or Bam HI (for enzyme I from strain H of B. amyloliquefaciens).ChemistryUse chemical symbols instead of words for elements, ions, or compounds, except at the beginning of a sentence. These symbols do not have to be defined the first time they are used. Where the representation is general and the chemical species is not specified, do not indicate the ionic charge (e.g., Ca, Fe, K, NH4, NO3, SO4, and PO4). Whenever a specific ion of known valence state is described in a manuscript, indicate the charge in superscripts as the charge number followed by a plus (+) or minus (−) sign; where the charge number is 1, use only the sign (e.g., Ca2+, NH4+, NO3−). Where the oxidation state is not obvious inCopyright © ASA–CSSA–SSSA, 5585 Guilford Rd., Madison, WI 53711, USA.a formula or where the oxidation state is known and is important, it should be designated by a roman numeral in parentheses; for example, Fe(II).The amounts and proportions of fertilizer nutrient elements must be expressed in terms of the elements or in other ways as needed for theoretical purposes. The amounts or proportions of the oxide forms (P2O5, K2O, etc.) may also be included, in parentheses.Give the full chemical names for compounds at first mention in the abstract or text. (If many names need mention, they may be listed in a table instead of parenthetically throughout the text.) E xamplEs:atrazine [6-chloro-N-ethyl-N′-(1-methylethyl)-1,3,5-triazine-2,4-diamine]cyanazine {2-[[4-chloro-6-(ethylamino)-1,3,5-triazin-2-yl] amino]-2-methyl p ro-panenitrile}If given in the abstract, the full chemical names do not need to be repeated in the text. Use the most up to date chemical names available. Thereafter, the common or generic name can be used (e.g., atrazine, 2,4-D, etc.). Trade names should be avoided whenever possible. If it is necessary to use a trade name, it should be capitalized and spelled out as specified by the trademark owner. Omit the various trademark symbols, such as ® and ™.In the United States and Canada, the authority for names of chemical compounds is Chemical Abstracts and its indexes. The American Chemical Society’s ACS Style Guide (Coghill and Garson, 2006) and the Council of Science Editors’ Scientific Style and Format (CSE, 2006) contain many additional details on nomenclature in chemistry and biochemistry. Publications of the American Chemical Society’s committee on nomencla-ture and the nomenclature commissions of the International Union of Pure and Applied Chemistry (IUPAC) are available through Chemical Abstracts Service, Columbus, OH.Chapter 7 of this book has further information regarding SI units and concentration.Information on pesticides and adjuvants is found in the Herbicide Handbook of the Weed Science Society of America (Ahrens, 1994), the Crop Protection Handbook (Meister,Table 3–2. Amino acids and their abbreviations.Amino acid Long abbreviation Short abbreviation Alanine Ala AArginine Arg R Asparagine Asn NAspartic acid Asp DCysteine Cys CGlutamic acid Glu E Glutamine Gln QGlycine Gly GHistidine His H Isoleucine Ile ILeucine Leu LLysine Lys K Methionine Met M Phenylalanine Phe FProline Pro PSerine Ser S Threonine Thr T Tryptophan Trp W Tyrosine Tyr YValine Val VCopyright © ASA–CSSA–SSSA, 5585 Guilford., Madison, WI 53711, USA.current edition), and the British Crop Protection Society’s Pesticide Manual: A Worldwide Compendium (Tomlin, 2011). See also the Merck Index (O'Neil, 2006, or current edition).The chemical names of the organic substances used for pesticides may include locants and descriptors consisting of numerals, letters (italic, roman, small-capital, or Greek letters), symbols, and words in various combinations. Below is a selection of com-mon usages:• Use italics for the prefixes anti, asym, c, cis, cyclo, d, endo, exo, l, m, n, o, p, r, s, sec, t, tert, and trans. Do not capitalize these prefixes, even at the beginning of a sentence or in a title.• Use italics for the capitalized prefixes R, R*, S, S*, E, and Z and enclose them in parentheses.• Use italics for symbols of chemical elements indicating ligation or attachment to an atom (e.g., O, P, N, S) or when indicating added hydrogen (H).• Use Greek letters to denote position or stereochemistry (e.g., a-amino acids). • Enclose the stereochemistry prefixes for plus and minus in parentheses: (+), (−), and (±).• Use roman (plain) type for multiplying prefixes (e.g., hemi, mono, di, tri, deca;semi, uni, sesqui, bi, ter, deci; bis, tris, decakis).For a full treatment with examples, including details of punctuation and capitalization in various contexts, see the ACS Style Guide (Coghill and Garson, 2006, Chapter 12).Copyright © ASA–CSSA–SSSA, 5585 Guilford Rd., Madison, WI 53711, USA.。
regularization paths for generalized linear models via coordinate descent
Stanford University
Trevor Hastie
Stanford University
Rob Tibshirani
Stanford University
Abstract We develop fast algorithms for estimation of generalized linear models with convex penalties. The models include linear regression, two-class logistic regression, and multinomial regression problems while the penalties include 1 (the lasso), 2 (ridge regression) and mixtures of the two (the elastic net). The algorithms use cyclical coordinate descent, computed along a regularization path. The methods can handle large problems and can also deal efficiently with sparse features. In comparative timings we find that the new algorithms are considerably faster than competing methods.
JSS
Journal of Statistical Software
January 2010, Volume 33, Issue 1. /
IBM Flex System Manager 产品指南说明书
IBM Flex System ManagerProduct Guide (withdrawn product)IBM® Flex System™ Manager (FSM) is a systems management appliance that drives efficiency and cost savings in the data center. IBM Flex System Manager provides a pre-integrated and virtualized management environment across servers, storage, and networking that is easily managed from a single interface. A single focus point for seamless multichassis management provides an instant and resource-oriented view of chassis and chassis resources for both IBM System x® and IBM Power Systems™compute nodes. You can reduce the number of interfaces, steps, and clicks it takes to manage IT resources, intelligently manage and deploy workloads based on resource availability and predefined policies, and manage events and alerts to increase system availability and reduce downtime while reducing operational costs.The IBM Flex System Manager management appliance is shown in the following figure.Figure 1. IBM Flex System Manager management applianceDid you know?Click here to check for updatesDid you know?IBM Flex System, a new category of computing and the next generation of Smarter Computing, offers intelligent workload deployment and management for maximum business agility. This chassis delivers high-speed performance complete with integrated servers, storage, and networking for multi-chassis management in data center compute environments. Furthermore, its flexible design can meet the needs of varying workloads with independently scalable IT resource pools for higher utilization and lower cost per workload. While increased security and resiliency protect vital information and promote maximum uptime, the integrated, easy-to-use management system reduces setup time and complexity, providing a quicker path to return on investment (ROI).IBM Flex System gives forward-thinking companies a way to completely rethink deployment and management of their IT environments; it offers an opportunity to evolve to a more open, agile, and integrated computing system that is dynamically managed from a single vantage point to simultaneously maximize efficiency and innovation.Figure 2. IBM Flex System Manager internal management network connectionsThe management network is used to complete management-related functions for the various endpoints that are managed by the IBM Flex System Manager management software, such as other Enterprise Chassis and compute nodes. During initialization, the management software discovers any Enterprise Chassis on the management network. The management node console can be connected to the management network or to the data network.The IBM Flex System Manager management node Eth1 interface (two 10 Gb Ethernet ports) must be connected to the chassis switch modules that are installed in I/O bay 1 or bay 2. This setup is referred to as the data network. You can configure a switch module in bay 1 or bay 2 to map Eth1 to one of its external Ethernet ports, as you would configure the other nodes in the chassis that are connected to the external network. The data network is used by applications and operating systems and can support data transfer rates up to 10 Gbps if a chassis switch module that is capable of 10 Gbps is installed.One of the key functions that the data network supports is discovery of operating systems on the various network endpoints. Discovery of operating systems by the IBM Flex System Manager is required to support software updates on an endpoint such as a compute node. The IBM Flex System Manager Checking and Updating Compute Nodes wizard assists you in discovering operating systems as part of the initial setup.Management task Agentlessin-band Agentlessout-of-bandPlatformAgentCommonAgentCommand Automation No No No Yes Hardware alerts and status No Yes Yes Yes Platform alerts No No Yes Yes Health and status monitoring No No Yes Yes File Transfer No No No Yes Inventory (hardware)No Yes Yes Yes Inventory (software)Yes No Yes Yes Process Management No No No Yes Power Management No Yes No Yes Remote Control No Yes No No Remote Command Line Yes No Yes Yes Resource Monitors Yes*No Yes Yes Update Manager (firmware)Yes**Yes Yes Yes Update Manager (agent updates)No No Yes Yes* Supported for VMware and Hyper-V virtualization environments** Supported for Windows environmentsThe following table shows supported virtualization environments and their management tasks.Table 6. Supported virtualization environments and management tasksVirtualization environment Management task AIX andLinux*IBM i VMware ESXwith vCenterMicrosoftHyper-VLinuxKVMDeploy virtual servers Yes Yes Yes Yes Yes Deploy virtual farms No No Yes No Yes Relocate virtual servers Yes Yes**Yes No Yes Maintenance mode Yes No Yes No Yes Import virtual appliance packages Yes Yes No No Yes Capture virtual servers Yes Yes No No Yes Capture workloads Yes Yes No No Yes Deploy virtual appliances Yes Yes No No Yes Deploy workloads Yes Yes No No Yes Deploy server system pools Yes Yes No No Yes Deploy storage system pools Yes Yes No No No* Linux on IBM Power Systems compute nodes** Supported only for virtual servers that are running IBM i v7.1, TR4 PTF group SF99707 level 4, or later. The following table shows supported I/O modules and their management tasks.Table 7. Supported I/O modules and management tasks (Part 1)I/O module Management task EN2092EN4091EN4093EN4093RCN4093SI4093Discovery Yes Yes Yes Yes Yes Inventory Yes Yes Yes Yes Yes Monitoring Yes Yes Yes Yes Yes Alerts Yes Yes Yes Yes Yes Protocol configuration Yes No Yes Yes No VLAN configuration Yes No Yes Yes No CEE configuration No No Yes Yes No EVB configuration No No Yes Yes No Stacked switch management No No Yes No NoTable 7. Supported I/O modules and management tasks (Part 2)I/O moduleManagement taskEN4023Cisco B22FC3171FC5022IB6131 Discovery Yes Yes Yes Yes Yes Inventory Yes Yes Yes Yes Yes Monitoring Yes Yes Yes Yes Yes Alerts Yes Yes Yes Yes Yes Protocol configuration No No Yes No No VLAN configuration No No No No No CEE configuration No No No No No EVB configuration No No No No No Stacked switch management No No No No NoThe following table shows supported virtual switches and their management tasks.Table 8. Supported virtual switches and management tasksVirtualization environment Linux KVM VMware vSphere PowerVM Hyper-V Virtual switchManagement taskPlatform Agent VMware IBM 5000V PowerVM Hyper-V Discovery Yes Yes Yes Yes No Inventory Yes Yes Yes Yes No Configuration management Yes Yes Yes Yes No Automated logical networkprovisioning (ALNP)Yes Yes Yes Yes NoThe following table shows supported storage systems and their management tasks.Table 9. Supported storage systems and management tasksTrademarksLenovo and the Lenovo logo are trademarks or registered trademarks of Lenovo in the United States, other countries, or both. A current list of Lenovo trademarks is available on the Web athttps:///us/en/legal/copytrade/.The following terms are trademarks of Lenovo in the United States, other countries, or both:Lenovo®Flex SystemSystem x®The following terms are trademarks of other companies:Intel® and Xeon® are trademarks of Intel Corporation or its subsidiaries.Linux® is the trademark of Linus Torvalds in the U.S. and other countries.Hyper-V®, Internet Explorer®, Microsoft®, and Windows® are trademarks of Microsoft Corporation in the United States, other countries, or both.Other company, product, or service names may be trademarks or service marks of others.。
A theory of firm scope
Objective Functions of d e c i s i o n - makers
• What should be t h e o b j e c t i v e function of d e c i s i o n - makers? • S o c ia l optimum: Decisions such t h a t t o t a l s u r p l u s va+wa+vb+ maximized
•13
Further Assumptions
·C o o r d i n a t i o n r a i s e s t o t a l monetary p r o f i t s but decreas l e a s t one unit manager‘s private benefits: benefits of coordination are unevenly distributed ·T o t a l s u r p l us maximized i f e i t h e r both p a r t i e s coordin i f neither does ·O n l y i f both u n i t s decide t o coordinate t h e payoffs cha from the non-coordination case (Assumption not necessary for qualitative results)
•12
Application : Horizontal Interaction
• Units have binary decision t o make: • Coordinate with other unit (Y) or d o n ‘ t coordinate (N)
maxim of relation例子 -回复
maxim of relation例子-回复什么是关系的“极大化准则”?极大化准则(maxim of relation)是德国哲学家奥托·冯·冯·提茨(Otto von Bismarck)于19世纪末提出的一条原则,用于解决人际关系和社交交往中的问题。
它指导人们在进行社交互动时,应尽力创造并维持积极和融洽的关系,避免冲突和敌意。
那么,如何应用极大化准则来处理人际关系呢?以下将一步一步回答。
首先,要意识到极大化准则的重要性。
有时候,人们可能习惯于争斗和挑战他人,而忽略了建立和谐关系的重要性。
然而,极大化准则告诉我们,积极的关系可以带来更多的好处,例如建立信任、获得支持和共同合作等。
其次,了解对方的需求和意愿。
人们彼此的需求和意愿可能不同,因此在建立关系时应尽量考虑对方的想法和感受。
了解他人的需求可以帮助我们更好地满足对方的期望,从而建立更加积极和融洽的关系。
第三,表达自己的意见和需求,同时尊重他人的观点。
在与他人交流时,我们应充分表达自己的想法和需求,让对方了解我们的立场。
然而,这并不意味着我们可以忽视他人的观点。
极大化准则告诉我们,要尊重他人的权利,包括他们的意见和观点。
只有在平等和开放的基础上进行交流,我们才能建立更加和谐的关系。
第四,掌握有效的沟通技巧。
沟通是建立和谐关系的重要因素。
通过良好的沟通,我们可以更好地理解他人的需求和意愿,并表达自己的观点和要求。
简洁明了、尊重对方、倾听对方的需求、避免冲突和非建设性的批评等,都是有效沟通的关键。
第五,培养共同利益和合作的意识。
合作是建立积极关系的重要方式之一。
通过合作,我们可以共同努力实现共同的目标和利益。
极大化准则告诉我们,积极的合作可以带来更多的好处,例如共同成就感、更大的资源和机会等。
第六,处理冲突和问题。
在人际关系中,难免会遇到冲突和问题。
然而,极大化准则要求我们以积极的态度解决问题,避免冲突的升级。
当面临冲突时,我们可以通过合理化对方的观点、提出解决方案、妥协和寻求中间地带等方式来解决问题。
MSlSS手册心理与教育测量
2001 VersionScott Huebner, Ph.D. University of South Carolina Department of Psychology Columbia, SC 29208Table of ContentsPages Introduction and Rationale (2)Scale Structure (3)Administration and Scoring (4)Normative Data (5)Reliability (5)Validity (5)Permission to Use (5)References .............................................................................................................................. 6-8Multidimensional Students’ Life Satisfaction Scale:Introduction and RationaleThe impetus for the construction of the Multidimensional Students Life Satisfaction Scale (MSLSS) was the increased interest in the promotion of positive psychological well-being in children and adolescents (Compass, 1993; Sarason, 1997). In contrast to models that infer well-being from the absence of psychopathological symptoms, the World Health Organization (1964) defined health as a state of complete physical, mental, and social well-being. Psychologists, such as Cowen (1991), shared this perspective, arguing that psychological well-being should be considered on the basis of positive indicators, including indicators like “a basic satisfaction with oneself and one’s existence…or life satisfaction” (p. 404).Life satisfaction has been defined as a “global evaluation by the person of his or her life” (Pavot, Diener, Colvin, & Sandvik, 1991, p. 150). Although hundreds of studies of life satisfaction of adults have been conducted (see Diener, 1994; Veenhoven, 1993), life satisfaction in childhood has only recently become the focus of empirical work. Recent investigations have demonstrated the incremental importance of t he life satisfaction construct in understanding children and adolescents’ psychological well-being. For example, life satisfaction reports have been differentiated from other well-being constructs such as self-esteem (Terry & Huebner, 1995; Lucas, Diener, & Suh, 1996; Huebner, Gilman, & Laughlin, 1999), depression (Lewinsohn, Redner, & Seely, 1991), positive affect (Lucas et al., 1996; Huebner, 1991c; Huebner, & Dew, 1996), and others.Systematic research has been hindered by the lack of well-validated instruments for children and adolescents (Bender, 1977; Huebner, 1997). To date, children’s life satisfaction instruments have been limited to unidimensional measures of global or general life satisfaction, which yield only a single overall score (e.g., Perceived Life Satisfaction Scale: Adelman, Taylor, & Nelson, 1989; Students’ Life Satisfaction Scale: Dew & Huebner, 1994; Huebner, 1991a & b).The MSLSS was designed to provide a multidimensional profile of children’s life satisfaction judgments. Such differentiated assessments are expected to enable more focused diagnostic, prevention, and intervention efforts. For example, students who indicate relatively high levels of dissatisfaction with their family experiences should necessitate different intervention strategies than students who indicate greater dissatisfaction with their school experiences. Such differentiated assessments may also yield more revealing comparisons with traditional objective indicators used to assess the quality of life of children and adolescents (e.g., divorce rates, family income levels, per pupil expenditures on schooling).Specifically, the MSLSS was designed to (a) provide a profile of children’s satisfaction with important, specific domains (e.g., school, family, friends) in their lives; (b) assess their general overall life satisfaction; (c) demonstrate acceptable psychometric properties (e.g., acceptable subscale reliability); (d) reveal a replicable factor structure indicating the meaningfulness of the five dimensions; and (e) be usedeffectively with children across a wide range of age (grades 3-12) and ability levels (e.g., children with mild developmental disabilities through gifted children).Scale StructureMSLSS Items FamilyI enjoy being at home with my family.My family gets along well together.I like spending time with my parents.My parents and I doing fun things together.My family is better than most.Members of my family talk nicely to one another.My parents treat me fairly.FriendsMy friends treat me well.My friends are nice to me.I wish I had different friends.*My friends are mean to me.*My friends are greatI have a bad time with my friends.*I have a lot of fun with my friends.I have enough friends.My friends will help me if I need it.SchoolI look forward to going to school.I like being in school.School is interesting.I wish I didn’t have to go to school.*There are many things about school I don’t like.*I enjoy school activities.I learn a lot at school.I feel bad at school.*Living EnvironmentI like where I live.I wish there were different people in my neighborhood.* I wish I lived in a different house.*I wish I lived somewhere else.*I like my neighborhood.I like my neighbors.This town is filled with mean people.*My family’s ho use is nice.There are lots of fun things to do where I live.Table 1 (continued)ItemsSelfI think I am good looking.I am fun to be around.I am a nice person.Most people like me.There are lots of things I can do well.I like to try new things.I like myself.*reverse keyed itemsAdministration and ScoringThe 40-item MSLSS may be administered to children in groups as well as individually. The instructions for the scale are provided prior to the rest of the scale, With younger children, (grades 3-5), it is recommended that the examiner read the directions aloud to the students and encourage them to ask questions as necessary. With all students, it is essential to monitor their responses to ensure that they respond appropriately (e.g., answer all questions, non-random and non-biased responding). The readability of the scale is at the 1.5 grade level, so most students require little or no assistance in responding to the questions.Scoring is straightforward. The four response options are assigned points as follows: (never = 1); (sometimes = 2); (often = 3); and (almost always = 4). Negatively-keyed items must be reverse scored (see pp. 3-4 for the list of negatively-keyed items). Hence, negatively-keyed items are scored so that almost always = 1, and so forth. Higher scores thus indicate higher levels of life satisfaction throughout the scale.It should be noted that a 6-point agreement format has been used with middle and high school students (Huebner et al., 1998). In this case, response options are assigned points as follows: (1 = strongly disagree, 2 = moderately disagree, etc.).Because the domains consist of unequal number of items, the domain and total scores are made comparable by summing the item responses and dividing by the number of domain (or total) items.Normative DataNormative data obtained to date are available for elementary (grades 3-5) (Huebner, 1994), middle (Huebner et al., 1998), and high school students (Gilman et al., 2000; Greenspoon & Saklofske, 2997; Huebner, 1994; Huebner, Laughlin, Ash, & Gilman, 1997).ReliabilityInternal consistency (alpha) coefficients have been reported in various publications (Dew, 1996; Greenspoon & Saklofske, 1997; Huebner, 1994; Huebner, Laughlin, Ash, & Gilman, 1997). The findings suggest that the reliabilities all range from .70s to low .90s; thus they are acceptable for research purposes. Test-retest coefficients for two- and four-week time periods have also been reported (Dew, 1996; Huebner et al., 1997; Huebner & Terry, 1995) falling mostly in the .70 - .90 range, providing further support for the reliability of the scale.ValidityThe results of exploratory factor analyses have supported the dimensionality of the MSLSS (Huebner, 1994). Confirmatory factor analyses have provided further support or the multidimensional, hierarchical model consisting of a general life satisfaction higher-order factor at the apex of the hierarchy along with five specific domains below (Gilman et al., 2000; Huebner et al., 1998). Findings have generalized to school age students in Canada (Greenspoon & Saklofske, 1997) Korea (Park, 2000), and Spain (Casas et al., 2000).Convergent and discriminant validity have also been demonstrated through predicted correlations with other self-report well-being indexes (Dew et al., 2001; Gilman et al., 2000; Greenspoon & Saklofske, 1997; Huebner, 1994; Huebner et al., 1998), parent reports (Dew et al., 2001; Gilman & Huebner, 1997), teacher reports (Huebner & Alderman, 1993), and social desirability scales (Huebner et al., 1998). Findings of weak relationships with demographic variables (e.g., age, gender) also fit with theoretical expectations (Huebner, 1994; Huebner et al., 1998).Nevertheless, additional validation research is needed to clarify the precise boundaries of the life satisfaction construct as well as the range of applications for particular children. For example, Ash and Huebner (1998) and Griffin and Huebner (2000) reported on unique aspects of the validity and usefulness of the MSLSS in the assessment of the well-being of two groups of exceptional children (i.e., academically gifted and emotionally disordered middle school students). Studies of the usefulness of the MSLSS and other life satisfaction scales with other groups of children (e.g., children with mental disabilities, ADHD) would be illuminating as well.Permission to UseThe MSLSS is in the public domain. Researchers may use it without permission. The author welcomes any feedback regarding its usefulness.ReferencesAdelman, H. S., Taylor, L., & Nelson, P. (1989). Minors’ dissatisfaction with their life circumstances. Child Psychiatry and Human Development, 20, 135-147.Ash C, & Huebner, E. S. (1998). Life satisfaction reports of gifted middle-school children. School Psychology Quarterly, 13, 310-321.Bender, T. A. (1997). Assessment of Subjective Well-Being During Childhood And Adolescence. In G. D. Phye (ed.) Handbook of classroom assessment: Learning, Achievement, and Adjustment (pp. 199-225). San Diego, CA: Academic Press.Casas, F., Alsinet, F., Rossich, M., Huebner, E. S., & Laughlin, J. (2000, July). Cross-cultural investigation of the Multidimensional Students’ Life Satisfaction Scale with Spanish adolescents. Paper presented at the Third Conference of International Quality of Life Studies, Girona, Spain.Compas, B. (1993). Promoting positive mental health in adolescence. In S. G. Millstein, A. C. Peterson, & E. O. Nightingale (Eds.), Promoting the health of adolescents (pp. 159-179). New York: Oxford University Press.Cowen, E. L. (1991). In pursuit of wellness. American Psychologist, 46, 404.408.Dew, T., Huebner, E. S., & Laughlin, J. E. (2001). The development and validation of a life satisfaction scale for adolescents. Manuscript submitted for publication.Dew, T. & Huebner, E. S. (1994). Adolescents perceived quality of life: An exploratory investigation. Journal of School Psychology 32, 185-199.Diener, E. (1994). Assessing subjective well-being: Progress and opportunities. Social Indicators Research, 31, 103-159.Gilman, R., Huebner, E. S., & Laughlin, J. (2000). A first study of the Multidimensional Students' Life Scale with adolescents. Social Indicators Research, 52, 135-160.Gilman, R., & Huebner, E. S. (1997). Children's reports of their well-being: Convergence across raters, time, and response formats. School Psychology International, 18, 229-243.Greenspoon, P. J. & Saklofske, D. H. (1997). Validity and reliability of the Multidimensional Students’ Life Satisfaction Scale with Canadian children. Journal of Psychoeducational Assessment, 15, 138-155.Griffin, M., & Huebner, E. S. (2000). Multidimensional life satisfaction reports of students with serious emotional disturbance. Journal of Psychoeducational Assessment, 18, 111-124.Huebner, E. S. (1991a). Initial development of the Students' Life Satisfaction Scale. School Psychology International, 12, 231-240.Huebner, E. S: (1991b). Correlates of life satisfaction in children. School Psychology Quarterly, 6, 103-111.Huebner, E. S. (1991c). Further validation of the Students' Life Satisfaction Scale: The independence of satisfaction and affect ratings. Journal of Psychoeducational Assessment, 9, 363-368.Huebner, E. S. (1994). Preliminary development and validation of a multidimensional life satisfaction scale for children. Psychological Assessment, 6,149-158.Huebner, E. S. (1997). Life satisfaction and happiness. In G. Bear, K. Minke, & A. Thomas (Eds.), Children's needs - II (pp. 271-278). Silver Spring, MD: National Association of School Psychologists.Huebner , E. S., & Alderman, G. L. (1993). Convergent and discriminant validation of a childrens' life satisfaction scale: Its relationship to self- and teacher-reported psychological problems and school functioning. Social Indicators Research, 30, 71-82.Huebner, E. S., & Dew, T. (1996). The interrelationships among life satisfaction, positive affect, and negative affect in an adolescent sample. Social Indicators Research, 38, 129-137.Huebner, E. S., Gilman, R. & Laughlin, J. (1999). The multidimensionality of children's well-being reports: Discriminant validity of life satisfaction and self-esteem. Social Indicators Research, 46, 1-22.Huebner, E. S., Laughlin, J. E., Ash C., & Gilman, R. (1998). Further validation of the Multidimensional Students' Life Satisfaction Scale. Journal of Psychological Assessment, 16, 118-134..Lewinsohn, P. M., Redner, E., & Seeley, J. R. (1991). The relationship between life satisfaction and psychosocial variables: New perspectives. In F. Strack, M. Argyle, & N. Schwarz (Eds.), Subjective well-being: An interdisciplinary perspective (pp. 193-212). New York: Pergamon.Lucas, R. E., Diener, E., & Suh, E. (1996). Discriminant validity of well-being measures. Journal of Personality and Social Psychology, 71, 616-628.Park, N. (2000). Life satisfaction of school age children: Cross-cultural and cross-developmental comparisons. Unpublished doctoral dissertation, University of South Carolina.Pavot, W., & Diener, E. (1993). Review of the Satisfaction With Life Scale. Psychological Assessment, 5, 164-172.Sarason, S. D. (1997). Forward. In R. Weissberg, T. P. Gullotta, R. L. Hampton, B. A. Ryan, & g. R. Adams (Eds.), Enhancing children’s wellness (Vol. 8) (p. ix-xi).。
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
Maximizing Stochastic Monotone Submodular Functions Arash Asadpour∗Hamid Nazerzadeh†Amin Saberi∗AbstractWe study the problem of maximizing a stochastic monotone submodular function with respect to a matroid constraint.We study the adaptivity gap-the ratio between the values of optimal adaptive andnon-adaptive policies-and show that it is equal to ee−1.This result implies that the benefit of adaptivityis bounded.We also study the myopic policy and show that it is a12-approximation.Furthermore,when thematroid is uniform,approximation ratio of the myopic policy becomes1−1ewhich is optimum.1IntroductionThe problem of maximizing submodular functions has been extensively studied in operations research and computer science.For a set A,the set function f:2A→R is submodular if for any two subsets S,T⊂A we havef(S∪T)+f(S∩T)≤f(S)+f(T)An equivalent definition is that the inequality below holds for any S⊆T⊆A and j∈Af(T+j)−f(T)≤f(S+j)−f(S)where f(·+j)denotes f(·∪{j}).Also,function f is monotone if for any two subsets S⊆T⊆A:f(S)≤f(T)A wide range of optimization problem that arise in the real world can be modeled as maximizing a monotone submodular functions with respect to some constraints.One instance is the welfare maximization problem[9,11,23]which is tofind an optimal allocation of resources to agents where the utilities of the agents are submodular.Submodularity corresponds to the law of diminishing return in economy.Another application of this problem is capital budgeting in which a risk-averse investor with a limited budget is interested infinding the optimal investment in different projects[24,2].The utility function of a risk averse investor is submodular.It is also naturally non-negative and monotone.Another example is the problem of viral marketing and maximizing influence through the network[14,18], where the goal is to choose an initial“active”set of people,so as to maximize the spread of a technology or behavior in a social network.It is well-known that under many models of influence propagation in networks (e.g.,cascade model[14]),the expected size of thefinal cascade is a submodular function of the set of initially activated individuals.Also,due to budget limitations,the number of people that we can activate in the beginning is bounded.Hence,the maximizing influence problem can be seen as a maximizing submodular function problem subject to cardinality constraints.Yet another example is the problem of optimal placement of sensors for environmental monitoring[16,17] where the objective is to place sensors in the environment in order to most effectively reduce uncertainty in observations.This problem can be modeled by entropy minimization and,due to the concavity of the entropy function,it is a special case of submodular optimization.∗Management Science and Engineering Department,Stanford University,Stanford,CA.{asadpour,saberi}@ †Microsoft Research,Cambridge,MA.hamidnz@1For the above problems and many others,the constraints can be model by a matroid.Afinite matroid M is defined by a pair(A,I),where I is a collection of subsets of A(called the independent sets)with the following properties:1.Every subset of an independent set is independent.2.If S and T are two independent sets and T has more elements than S,then there exists an element inT which is not in S and when added to S still gives an independent set.Two important special cases are uniform matroid and partition matroid.In a uniform matroid,all the subsets of A of size at most k,for a given k,are independent.Uniform matroids represent cardinality constraints.A partition matroid is defined over a partition of set A,where every independent set includes at most one element from each set in the partition.The celebrated result of Nemhauser et al.[20]shows that for maximizing nonnegative monotone sub-modular functions over uniform matroids,the greedy algorithm gives a(1−1e ≈0.632)-approximation ofthe optimal ter,they showed that for optimizing over matroids,the approximation ratio of thegreedy algorithm is12.Recently,Calinescu et al[5]proposed a better approximation algorithm with ratio1−1e .It also has been shown that this factor is optimal(in the value oracle model),if only a polynomialnumber of queries is allowed[19,10].However,these algorithms are designed for deterministic environments.In practice,one must deal with the stochasticity caused by the uncertain nature of the problem,the incomplete information about the environment,etc.For instance,in welfare maximization,the quality of the resources may be unknown in advance,or in the capital budgeting problem some projects taken by an investor may fail due to unexpected events in the market.As another example,in viral marketing some people in the initial set might not adopt the behavior.Also,in the environmental monitoring example,it is expected that a non-negligible fraction of sensors might not work properly for various reasons.All these possibilities motivate the problem of stochastic submodular maximization.In the stochastic setting,the outcome of the elements in the selected set are not known in advance and they will be only discovered after they are chosen.1.1Problem DefinitionIn defining the problem,we need to use some care to maintain generality.Consider a set A={X1,···,X n} of n independent random variables over a domain∆.The domain varies depending on the application.For instance,in the welfare maximization problem,X i denotes the quality of the resource or in viral marketing X i corresponds to the set of people who are influenced by person i.The distribution of each X i is potentially different and is given by a function g i.Let x i denote a realization of X i.Also,let vector s=<ˆx1,···,ˆx n>denote a realization of set S⊂A, whereˆx i=x i for X i∈S andˆx i=0for i/∈S.For a given function f:∆n→R+,we can define the stochastic function F:A→R+as F(S)=E[f(s)],where s is a realization of S and the expectation is taken with respect to the product distribution defined by g i’s.Also,consider a subset T∈A,and a realization t of T.We can define a conditional expectation E[f(s)|t]. In the distribution imposed by conditioning on t,s i=t i if its corresponding random variable is in S∩T. Otherwise s i is chosen independently with respect to the distribution defined by g i’s.Let us denote this conditional expectation with F(S,t).We call the set function F stochastic monotone submodular if F(·,t)is monotone submodular for every t. Observe that if f is monotone submodular then F is stochastic monotone submodular because it is a convex combination of monotone submodular functions.Remark:We assume that either we can compute the value of function F up to a desired degree of accuracy explicitly,or F is given to us via an“oracle”.This is a natural assumption for all the applications mentioned in the paper.In fact,in most cases the expectations can be computed simply by using sampling.For example,sampling works when the probability distribution functions are constant Lipschitz continuous,or2when their support is a polynomial size set of discrete values.In both cases,a small o(1)error is introduced in the calculations that we ignore in the rest of the paper.Definition:[Maximizing a stochastic monotone submodular function]The set A={X1,···,X n}of n independent random variables,the matroid M=(A,I),and the stochastic monotone submodular set function F:2A→R+are given.Find a subset S∈I that maximizes F,i.e.,max S∈I E[F(S)]where the expectation is taken over the probability distribution of the sets chosen by the policy.A special case of the above problem is the stochastic max k-cover problem which is defined as follows. Suppose a collection A of random subsets of N={1,2,···,m}are given.Each element X i∈A is a random subset of N,and it is distribution is denoted by a probability distribution g i.In the stochastic maximum k-cover problem,the goal is to choose k elements of A such that their union has the maximum cardinality. We discuss this problem in more detail in Section2.1.For the problem of maximizing a stochastic monotone submodular function,we study two types of policies: adaptive and non-adaptive.A non-adaptive policy is represented by afixed subset of A.An adaptive policy is a decision tree.It assumes that the value of each random variable can be observed as soon as it is chosen and it uses the observed values of the previously chosen elements to determine the next element in the subset.We compare these policies by studying the adaptivity gap of the problem.The adaptivity gap is defined as the ratio between the expected values of optimal adaptive and non-adaptive policies.Adaptivity gap has been previously studied for stochastic maximization problems with respect to covering[13]and packing[7,8] constraints.1.2ResultsWe present approximately optimal policies for the stochastic monotone submodular maximization problem. First,in Section2,we compare the performance of the optimal adaptive and non-adaptive policies.Although non-adaptive policies may not perform as well as adaptive ones,they are particularly useful when it is difficult or time consuming to discover the outcome of an element.For example,in the capital budgeting problem, it is not possible for the investor to wait until the end of each project to measure the success,or in the environmental monitoring problem,it is not practical to measure the performance of sensors after placing each sensor in the environment.Surprisingly,we learn that the adaptivity gap of the problem is equal to ee−1≈1.59.In other words,there exists a non-adaptive policy which achieves at least e−1e fraction of the value of best adaptive policy.This result leads to a(e−1e )2≈40%approximation of the optimal adaptive policy by a non-adaptive policythat runs in polynomial time in n.We also give an example to show that our analysis of the adaptivity gap is tight.For that,we use a simple instance of the stochastic max k-cover problem.In Section3,we focus on natural myopic policies.We study the natural extension of the myopic policy studied in[5]in a stochastic environment.This policy iteratively chooses an element with the maximum expected marginal value,conditioned on the outcome of the previous elements.We show that the approximation ratio of this policy with respect to the optimal adaptive policy is12forgeneral matroids.We also prove that over uniform matroid(i.e.,subject to a cardinality constraint),theapproximation ratio of this policy is1−1e .1Due to the results of[19,10],the approximation ratio of1−1eis optimal only if a polynomial number of oracle accesses is allowed.The closest work to ours in the literature is by Chan and Farias[4].They mainly study the problem of stochastic submodular optimization over partition matroids.In their model,there is an ordering over the partitions and any adaptive policy has to choose one element from each partition according to the givenorder.They present a12-approximation of the optimal adaptive policy(that respects the ordering)using amyopic policy.In our setting,we do not have afixed ordering.In addition,we prove most of our results for general matroids.1The results for the uniform matroid has appeared in a preliminary version of this work[3].32The Adaptivity Gap of Stochastic Submodular Optimization ProblemIn this section,we analyze the optimal adaptive and non-adaptive policies and compare the performance of the two.First,observe that since non-adaptive policies do not observe the realized value of the items until the end,they may choose all the elements in one step.In other words,any non-adaptive policy can be represented by the set of chosen elements.On the other hand,an optimal adaptive policy selects the elements based on the realized values of the previously chosen elements.Note that the policy knows the probability distribution of the values of the elements that are not yet chosen,but not their actual values.Although an adaptive policy can clearly perform better than a non-adaptive policy,we show that its advantage is limited.The main result of this section is as follows:Theorem1The adaptivity gap of the stochastic monotone submodular maximization problem is equal to ee−1.In order to prove the above theorem,we start by establishing an upper bound on the adaptivity gap.In Section2.1,we give an example that shows our analysis of the adaptivity gap is tight.Before proving the theorem,observe that since F(S)is a submodular function,we can use the following result of Calinescu et.al.[5]:Theorem2(Calinescu et al[5])Given oracle access to F(see Remark1),there exists a polynomial time algorithm that achieves an approximation ratio of1−1e−o(1).The above theorem immediately implies thatCorollary3A(1−1e −o(1))-approximation of the optimal non-adaptive policy can be computed in polynomialtime.Theorem1and the above corollary imply that:Corollary4There is a policy that is non-adaptive and also runs in polynomial time and computes a solution that is within(e−1e)2of the optimal adaptive policy.In the rest of this section,we prove Theorem1.The proof is inspired by the techniques developed in Section3.5of[21]for submodular optimization(in a non-stochastic setting).For the sake of consistency,we use the same notation as[21]wherever possible.We start by making a few observations about adaptive policies.First,any adaptive policy can be described by a(possibly randomized)decision tree in which at each step an element is being added to the current selection.Consider an arbitrary adaptive policy Adapt.Each path from the root to a leaf of this tree corresponds to a realization s∈ˆI of the sequence of elements chosen by Adapt.Here,ˆI denotes the set of all possible realizations of sets in I.Let y=<y1,···,y n>represent the probability that each element of A is chosen by Adapt,i.e.,y i is the probability of choosing X i.These probabilities sum up to1.Also, letβs denote the probability density function for outcome s∈ˆI.Then,we have the following properties:1.s∈ˆIβs=1.2.∀s:βs≥0.3.∀i,dx:s,s i∈dx iβs ds=y i g i(x i)dx i4Thefirst two properties hold becauseβdefines a probability measure on the space of all feasible outcomes. The third property implies that the probability that we observe outcome x i(a realized value of X i)among all possible outcome s is equal to the probability that X i is chosen(i.e.,y i)multiplied by the probability that the outcome is equal to x i.This property holds because of the independence among the random variables.Since every policy satisfies the above properties,we can establish an upper bound on the value of any adaptive policy.Hence,we define the function f+:[0,1]n→R as follows:f+(y)=supαs∈ˆIαs f(s):sαs=1,αs≥0,∀i,dx i:s,s i∈dx iαs ds=y i g i(x i)dx i.(1)Another observation is that for an optimal adaptive policy,vector y described above is in the base polytope of M(defined as follows).A set S∈I is called a basis for the matroid if|S|=max{|T|:T∈I}. The base polytope,B(M),is defined as:B(M)=conv{1S|S∈I,S is a basis}Here“conv”denotes the convex hull and1S is the characteristic vector of S,i.e.,1for elements in S and 0for other elements.Lemma5The expected value of the optimal adaptive policy is at most max y∈B(M){f+(y)},Proof:Note that an optimal adaptive policy only chooses independent sets.Due to monotonicity,all of these are independent sets are bases of the matroid.Hence,for an optimal adaptive policy vector y defined above is in B(M).Moreover,the expected value of the adaptive policy is bounded by f+(y),because the policy has to satisfy the3properties mentioned earlier.Now,we define an extension of set function F(S)to the domain of real numbers.For vector y∈[0,1]n, let Y denoted a random set where Y includes X i∈A with probability y i.With abuse of notation,we define the extension F:[0,1]n→R+as follows:F(y)=E[F(Y)]=Y is a basis of Ii∈Yy ii/∈Y(1−y i)F(Y).Function f+(y)sets an upper bound on the adaptive policies.We now establish a lower bound on the value of optimal non-adaptive policies via the following lemma from[21](Lemma3.4),which is based on pipage rounding[1].Lemma6[21]Any vector y∈B(M)can be rounded to an integral solution S∈I of value F(S)≥F(y).To complete the proof we need to show that for any vector y,the values of F(y)and f+(y)are within a constant factor of each other,which is established by combining Lemmas3.7and3.8from[21].Lemma7[21]For any monotone submodular function f and any vector y we havef+(y)≤(ee−1)F(y)Proof:[Theorem1]Lemma5shows that max y∈B(M)f+(y)is an upper-bound on the performance of the optimal adaptive policy.Consider y∗∈argmax y∈B(M)f+(y).By Lemma7,we have F(y∗)is at least a(1−1e )fraction of the expected value of an optimal adaptive policy.On the other hand,Lemma6impliesthat there exists a S∈I such that F(S)≥F(y∗).Note that F(S)is in fact the expected value gained by anon-adaptive policy that selects set S.Hence,S is a(1−1e )-approximation of the optimal adaptive policy.By Proposition10in the next section,this factor is tight.52.1A Tight Example:Stochastic Maximum k -CoverGiven a collection A of the subsets of N ={1,2,···,n },the goal of the max k -cover problem is to find k subsets from A such that their union has the maximum cardinality [10].In the stochastic version,the subset that an element of A would cover is revealed only after choosing the element,according to a given probability distribution.The following reduction shows that this problem is a special case of the stochastic monotone submodular maximization.For S ∈A ,let F (S )denote the expected number of elements covered by the subsets in S .Clearly,F is monotone.Consider two subsets S ⊆T ⊆A ,an element X ∈A ,and a realization y of an arbitrary subset of A .Because ∪A ∈S A ⊆∪B ∈T B ,for every realization y ,we have F (S +X )−F (S )≥F (T +X )−F (T ).In addition,M =(A ,{S ⊆A :|S |≤k })forms a uniform matroid.Therefore,the stochastic max k -cover problem is in fact a stochastic monotone submodular maximization problem.In this section,we define an instance of stochastic max k -cover problem that gives a lower bound on the adaptivity gap.This example has been brought to our attention by Vondr´a k [22].Consider the following instance:a ground set N ={1,2,···,n }and a collection A ={X (i )j |1≤i ≤n,1≤j ≤n 2}of its subsets are given.For every i,j ,define X (i )j to be the one-element subset {i }withprobability 1n and the empty set with probability 1−1n .The goal is to cover the maximum number of the elements of N by selecting at most k =n 2subsets from A .Lemma 8The optimal non-adaptive policy is to pick n subsets from each of the collections A (i )={X (i )j |1≤j ≤n 2}for every i .For large enough values of n ,the expected value of this policy is (arbitrarily close to)(1−1e )n .Proof :Consider an arbitrary non-adaptive policy which picks S ,containing n 2sets from A .For each i ,define k i =|S ∩A (i )|.Moreover,each element i ∈N is covered if and only if at least one of its correspondingchosen subsets are realized as a non-empty subset.Hence,it willbe covered with probability 1−(1−1n )k i .Therefore,the expected value of this policyis i 1−(1−1n )k i .Note that 1−(1−1n )x is a concave function with respect to x ,and also i k i =n 2.Hence,the expected value of the policy is maximized whenk 1=k 2=···=k n =n .In this case,the expected value is (1−(1−1n )n )n ≈(1−1e )n for large n .We now consider the following myopic adaptive policy P :Start with i =1and pick the elements of A (i )one by one until one of them is realized as {i }or all of elements in A (i )are chosen.Then increase i by one.Continue the iteration untill i =n +1.The following lemma gives a lower bound on the number of elements in N covered by the adaptive policy.Lemma 9The expected number of elements in N covered by P described above is (1−o (1))n .Proof :Let X k be the indicator random variable corresponding to the event that the subset chosen at the k -th step is realized as a non-empty subset for any 1≤k ≤n 2.Note that the number of elements covered by P is n 2k =1X k .Moreover,all X k ’s are independent random variables.By the description of P ,as long as k i =1X k <n 2,X k will be one with probability 1n and will be zero with probability 1−1n .Also,when t i =1X k =n ,we have already covered all the elements in N .Therefore,X t +1,···,X n 2will all be equal to zero.With this observation,we define i.i.d random variablesY 1,Y 2,···,Y n 2,where each Y i is set to be one with probability 1n and zero with probability 1n .Observe thatmin {n,Y = k Y k }has the same probability distribution as k X k .Note that E[Y ]=n .Using Chernoffbound,we havePr[Y ≤n −n 2/3]≤e −n 4/32n =e −n 1/3.Thus,with probability at least 1−e −n1/3we have Y >n −n 2/3.Hence,E[n 2 k =1X k ]=E[min {n,Y }]≥(1−e −n 1/3)(n −n 2/3)=n −o (n ),6The myopic adaptive policy:Initialize t=0,S0=∅,U0=∅While(A=U t∪S t)t←t+1S t←S t−1RepeatSelect X i∈argmax Xi ∈A\(U t−1∪S t−1)E[F(S t−1+X i)|s t−1]If S t−1∪{X i}/∈I thenU t←U t−1∪{X i}elseS t←S t−1∪{X i}U t←U t−1Observe x i and update s tUntil(A=U t−1∪S t−1)or(S t=S t−1)which completes the proof of the lemma.By combining the results of Lemmas8and9we have the following proposition:Proposition10For large enough n,the adaptivity gap of stochastic maximum coverage is at least ee−1.3Approximation Ratio of Simple Myopic PoliciesIn this section,we present an adaptive myopic policy with an approximation ratio of12with respect to anoptimal adaptive policy.In Section3.1,we show that the myopic policy achieves the approximation ratio of1−1e if the matroid is uniform.Note that even if the actual values were known,the problem of computingthe optimal policy is intractable.As mentioned before,the maximum k-cover is a special case of our problemand Feige[10]has shown that it is not possible tofind an approximation ratio better than1−1e for themaximum k-cover problem,unless NP⊂T IME(n O(log log n)).The policy is given in the abovefigure.At each iteration,from the elements in A that are not yet considered,the policy chooses an element with the maximum expected marginal value.We denote by S t the set of elements chosen by the adaptive policy up to iteration t.Let s t denote the realization of all these elements.Also,U t is the set of elements considered but not chosen by the policy due to the matroid constraint.Here is the main result of this section.Theorem11For general matroids,the approximation ratio of the myopic adaptive policy with respect to any optimal adaptive policy is12.Define∆t=F(S t)−F(S t−1).Also,let k be the number of elements chosen by the myopic policy(which is simply the rank of the matroid M).The basic idea of the proof is similar to Fisher et al.[12].But, the main difficulty is that the realized values of∆t are not always decreasing(due to the stochastic nature of the problem).In addition,the sequence of elements chosen by the optimal adaptive policy is random.7However,E[∆t|s t−1]≥E[∆t+1|s t−1](Note that E[∆t|s t]≥E[∆t+1|s t]does not necessary hold).Based on this observation,we prove the theorem.We will also use the following lemma from[12].Lemma12[12]For t=1,···,k,we have ti=1|C i|≤t.Note that T,U,and S are random sets,but the lemma holds for every realization because it is a consequence of the matroid constraint,not the realizations of the element chosen by the policy.We are now ready to prove the theorem.Proof:[Theorem11]Let P be the(random)set of elements chosen by the optimal adaptive policy. Also,for t=1,···,k,define C t=P∩(U t+1\U t).Consider a realization s t of S t.Because F is stochastic monotone submodular we haveE[F(P)|s t]≤E[l∈P\SF(S+l)|s t]The expectations,and in the rest of the proof,are taken over the probability distribution of all realizations of P such that the realized values of elements in P∩S t are according to s t.Since the above inequality holds for all s t,we haveE[F(P)]≤E[l∈P\SF(S+l)]E[F(P)]−E[F(S)]≤E[l∈P\S(F(S+l)−F(S))]Note that kt=1C t=P\S.Hence,E[F(P)]−E[F(S)]≤kt=1E[l∈C t(F(S+l)−F(S))]By expanding the expectation we haveE[F(P)]−E[F(S)]≤kt=1s t−1:S t−1∈IE[l∈C tF(S+l)−F(S)|s t−1]Pr[s t−1]ds t−1(2)Observe that conditioned on s t−1,because the myopic policy chooses an element with the maximum marginal value,we have∆t≥F(S+l)−F(S),l∈C t.Therefore,E[l∈C t F(S+l)−F(S)|s t−1]≤E[l∈C t∆t|s t−1]By plugging the above inequality into(2),we getE[F(P)]−E[F(S)]≤kt=1s t−1:S t−1∈IE[l∈C t∆t|s t−1]Pr[s t−1]ds t−1Using telescopic sums and the linearity of expectation we derive the following.Here∆k+1is defined0.E[F(P)]−E[F(S)]≤kt=1s t−1:S t−1∈IE[l∈C tkj=t(∆j−∆j+1)|s t−1]Pr[s t−1]ds t−1=kj=1jt=1s t−1:S t−1∈IE[l∈C t(∆j−∆j+1)|s t−1]Pr[s t−1]ds t−18Note that by using the Bayes’theorem and the law of total probability,for every t and j the integral term in the above is in fact equal to E[ l ∈C t (∆j −∆j +1)].Now,we can change the probability measure to calculate this expectation from s t −1to s j −1.Hence,we haveE[F (P )]−E[F (S )]≤kj =1j t =1 s j −1:S j −1∈I E[ l ∈C t(∆j −∆j +1)|s j −1]Pr[s j −1]ds j −1=kj =1j t =1 s j −1:S j −1∈I (E [|C t |E[∆j −∆j +1|s j −1]|s j −1])Pr[s j −1]ds j −1Note that conditioned on s j −1,the term E[∆j −∆j +1|s j −1]is by definition a constant and we can take it out from the outer expectation.Hence,E[F (P )]−E[F (S )]≤k j =1 s j −1:S j −1∈I E[j t =1|C t ||s j −1]E[∆j −∆j +1|s j −1] Pr[s j −1]ds j −1We now use Lemma 12which implies that in every realization j t =1|C t |≤j .We also use the fact that dueto the submodularity and the rule of the policy,we have E[(∆j −∆j +1)|s j −1]≥0.We conclude thatE[F (P )]−E[F (S )]≤kj =1 s j −1:S j −1∈I j E[(∆j −∆j +1)|s j −1]Pr[s j −1]ds j −1=kj =1 s j −1:S j −1∈I E[∆j |s j −1]Pr[s j −1]ds j −1=k j =1E[∆j ]=E[F (S )]Therefore,E[F (P )]≤2E[F (S )],as desired.Fisher et al.[12]have shown that even in the non-stochastic setting,in the worst-case,the approximation ratio of the greedy algorithm (hence the myopic policy)is equal to 12.Also,it is easy to see that that if Mis an intersection of κmatroids,then the approximation ratio of the myopic policy is equal to 11+κ.3.1Uniform MatroidsIn this section we show that the myopic policy described in the previous section has a better approximation ratio if the matroid is uniform.Theorem 13Consider the adaptive myopic policy that at each step selects an element with the maximum marginal value,conditioned on the realized value of the previously chosen elements.Over uniform matroids,the approximation ratio of this policy compared to the optimal adaptive policy is 1−1e .The proof presented here is similar to the proof of Kleinberg et al.[15]for submodular set functions.The main technical difficulty in our case is that the optimal adaptive policy here is a random set whose distribution depends on the realized values of the elements of A .Proof :Let P denote the (random)set chosen by an optimal adaptive policy.Also,denote the marginal value of the t -th element chosen by the myopic policy by ∆t ,i.e.,∆t =F (S t )−F (S t −1)9。