Dual Coordinate Descent Algorithms for Efficient Large Margin有效大幅度的双坐标下降算法.pptxVIP

下载本文档

32
0
约4.74千字
约 29页
2017-06-15 发布于上海
举报
版权申诉

Dual Coordinate Descent Algorithms for Efficient Large Margin有效大幅度的双坐标下降算法.pptx

1、有哪些信誉好的足球投注网站（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Dual Coordinate Descent Algorithms for EfficientLarge Margin Structured Prediction Ming-Wei Chang and Scott Wen-tau Yih Microsoft Research 1 Motivation Many NLP tasks are structured Parsing, Coreference, Chunking, SRL, Summarization, Machine translation, Entity Linking,… Inference is required Find the structure with the best score according to the model Goal: a better/faster linear structured learning algorithm Using Structural SVM What can be done for perceptron? 2 Two key parts of Structured Prediction Common training procedure (algorithm perspective) Perceptron: Inference and Update procedures are coupled Inference is expensive But we only use the result once in a fixed step 3 Observations 4 Observations Inference and Update procedures can be decoupled If we cache inference results/structures Advantage Better balance (e.g. more updating; less inference) Need to do this carefully… We still need inference at test time Need to control the algorithm such that it converges 5 Questions Can we guarantee the convergence of the algorithm? Can we control the cache such that it is not too large? Is the balanced approach better than the “coupled” one? 6 Contributions We propose a Dual Coordinate Descent (DCD) Algorithm For L2-Loss Structural SVM; Most people solve L1-Loss SSVM DCD decouples Inference and Update procedures Easy to implement; Enables “inference-less” learning Results Competitive to online learning algorithms; Guarantee to converge [Optimization] DCD algorithms are faster than cutting plane/ SGD Balance control makes the algorithm converges faster (in practice) Myth Structural SVM is slower than Perceptron 7 Outline Structured SVM Background Dual Formulations Dual Coordinate Descent Algorithm Hybrid-Style Algorithm Experiments Other possibilities 8 Structured Learning 9 The Perceptron Algorithm 10 Structural SVM Objective function Distance-Augmented Argmax 11 Dual formulation 12 Outline Structured SVM Background Dual Formu