2014-6-28
收藏了 //@ml_fans: //@我爱机器学习:还有@刘知远THU 老师提到的2007 Science的文章:Clustering by Passing Messages Between Data Points（PDF: http://genes.toronto.edu/affinitypropagation/FreyDueckScience07.pdf ） //@我爱机器学习:汇总下，2006年Hinton在Science上的论文：Reducing the dimensionality of data with neural network @阿饭AKA_Charles-Lao 最新一期的science上发了一篇关于clustering的文章，简洁易懂，example里面200行多的code，拜服。http://www.sciencemag.org/content/344/6191/1492.abstract
2014-6-29
http://jasperproject.github.io/ jasperproject Jasper is an open source platform for developing always-on, voice-controlled applications
http://mi.eng.cam.ac.uk/~mh521/ Matthew Henderson Statistical Dialogue Systems PhD Student, University of Cambridge
My PhD is supported by a Google Fellowship in Speech Technology. I am currently the lead organiser for the second and third Dialog State Tracking Challenges, research challenges on large offline corpora focused on improving the state of the art in tracking the state of spoken。 @tony的微博 O网页链接 Matthew Henderson Statistical Dialogue Systems PhD Student, University of Cambridge
Word-based Dialog State Tracking with Recurrent Neural Networks Deep Neural Network Approach for the Dialog State Tracking Challenge Discriminative Spoken Language Understanding Using Word Confusion Networks @tony的微博 O网页链接 Matthew Henderson Statistical Dialogue Systems PhD Student, University of Cambridge
http://mi.eng.cam.ac.uk/~sjy/publications.html Steve Young many papers about spoken dialogue system Stochastic Language Generation in Dialogue using Factored Language Models Word-Based Dialog State Tracking with Recurrent Neural Networks Deep Neural Network Approach for the Dialog State Tracking Challenge
http://research.microsoft.com/en-us/people/jawillia/ Jason Williams •Dialog management: Applications of machine learning; integration of expert knowledge; multi-modal dialog management; planning techniques; on-line improvement •Dialog and user modeling:
•Dialog and user modeling: Tracking and quantifying uncertainty in dialog state for human/computer dialog; representational structures for dialog state; ontology integration; simulation @tony的微博 O网页链接 Jason Williams •Dialog management: Applications of machine learning; integration of expert knowledge; multi-modal dialog management; planning techniques; on-line improvement •Dialog and user modeling:
•Planning under uncertainty: Markov decision processes (MDPs); partially observable Markov decision processes (POMDPs); reinforcement learning. @tony的微博 O网页链接 Jason Williams •Dialog management: Applications of machine learning; integration of expert knowledge; multi-modal dialog management; planning techniques; on-line improvement •Dialog and user modeling:
2014-6-30
deep architecture来做短文本匹配，这个很有想象力！ @高斌MS 华为诺亚方舟实验室吕正东研究员正在做invited talk，内容是用deep architecture来做短文本匹配。 2北四环中路辅路
http://papers.nips.cc/paper/5019-a-deep-architecture-for-matching-short-texts.pdf A Deep Architecture for Matching Short Texts //@tony的微博:deep architecture来做短文本匹配，这个很有想象力！ @高斌MS 华为诺亚方舟实验室吕正东研究员正在做invited talk，内容是用deep architecture来做短文本匹配。 2北四环中路辅路
http://smir2014.noahlab.com.hk/paper%204.pdf Deep Semantic Embedding //@tony的微博:deep architecture来做短文本匹配，这个很有想象力！ @高斌MS 华为诺亚方舟实验室吕正东研究员正在做invited talk，内容是用deep architecture来做短文本匹配。 2北四环中路辅路
2014-7-1
https://www.cse.ust.hk/pg/seminars/S14/li.html Semantic Matching: The Next Big Thing for Natural Language Processing?” Speaker: Dr. Hang LI Chief Scientist, Noah's Ark Lab at Huawei
introduce our recent work on using machine learning techniques to construct models for semantic matching. These include latent space model for query document matching in search, string re-writing kernel for question answering, and deep matching model for short text conversation. @tony的微博 O网页链接 Semantic Matching: The Next Big Thing for Natural Language Processing?” Speaker: Dr. Hang LI Chief Scientist, Noah's Ark Lab at Huawei
Abstract:
Most natural language processing (NLP) tasks, such as information retrieval, question answering, and machine translation, are based on matching between language expressions. This approach works quite well in practice; its limitation is also obvious, however. Sometimes mismatch between language expressions can occur. We argue that ‘semantic matching’ is an effective approach to overcome the challenge, that is to conduct more semantic analysis and perform matching between language expressions at semantic level. In this talk, I will first point out why semantic matching can help significantly enhance the performance of NLP. I will then justify my argument with some examples. More specifically, I will introduce our recent work on using machine learning techniques to construct models for semantic matching. These include latent space model for query document matching in search, string re-writing kernel for question answering, and deep matching model for short text conversation.
Biography:
Hang LI is chief scientist of the Noah’s Ark Lab at Huawei. He is also adjunct professor of Peking University and Nanjing University. His research areas include information retrieval, natural language processing, statistical machine learning, and data mining. He graduated from Kyoto University in 1988 and earned his PhD from the University of Tokyo in 1998. He worked at the NEC lab in Japan during 1991 and 2001, and Microsoft Research Asia during 2001 and 2012. He joined Huawei Technologies in 2012. Hang has more than 100 publications at top international journals and conferences, including SIGIR, WWW, WSDM, ACL EMNLP, ICML, NIPS, and SIGKDD. He and his colleagues’ papers received the SIGKDD’08 best application paper award, the SIGIR’08 best student paper award, and the ACL’12 best student paper award. Hang has also been working on the development of several products. These include Microsoft SQL Server 2005, Microsoft Office 2007 and Office 2010, Microsoft Live Search 2008, Microsoft Bing 2009 and Bing 2010. He has also been very active in the research communities and served or is serving the top conferences and journals. Recently, he is senior program committee members or area chairs of WSDM'13, IJCAI’13, KDD’13, ACL’13, ACML’13, and EMNLP’13; program committee members of WWW’13, AIRS’13, NIPS’13, and ICDM’13; demo co-chair of IJCNLP’13, and editorial board members of Computational Linguistics, Journal of the American Society for Information Science, ACM Transaction on Intelligent Systems and Technology, and the Journal of Computer Science & Technology.
http://www.textkernel.com/2014/05/deep-learning-technology-in-cv-parsing/ textkernel.com Innovation in CV parsing with Deep Learning Technology
Deep Learning is the next big thing in language technology and is considered to be the future of Artificial Intelligence. Deep Learning is a set of algorithms used on large amounts of data to automatically learn how to represent information, such as learning the meaning of new wo @tony的微博 O网页链接 textkernel.com Innovation in CV parsing with Deep Learning Technology
Textkernel specialises in multilingual semantic recruitment technology and provides recruiting tools to accelerate the process of matching demand and supply in the job market: multi-lingual CV parsing (available in 15 languages), job parsing and semantic searching, sourcing and @tony的微博 O网页链接 textkernel.com Innovation in CV parsing with Deep Learning Technology
matching short text，诺亚方舟实验室目前好像做了不少这方面的工作。 //@李航博士:1. 互联网搜索查询数据 http://blog.sina.com.cn/s/blog_7ad48fee010110o3.html 2. Paraphrase数据 http://research.microsoft.com/en-us/downloads/607d14d9-20cd-47e3-85bc-a2f65cd28042/ 3. 短文本对话数据 http://data.noahlab.com.hk/conversation/ @龙星镖局 帮同学问下：哪里有可用于matching short text (如相似句子，相似query)的 data set？@李航博士 @数据堂
http://smir2014.noahlab.com.hk/SMIR2014.htm The main purpose of the workshop SMIR 2014 is to bring together IR and NLP researchers working on or interested in semantic matching, to share latest research results, express opinions on the related issues, and discuss future directions.
SM can be extended to phrases or sentences. Indeed, there are initiatives such as the semantic textual similarity (STS) evaluation campaign, which go beyond term level matching and aim at capturing the semantic relations between entire phrases as well as those between entire sent @tony的微博 O网页链接 The main purpose of the workshop SMIR 2014 is to bring together IR and NLP researchers working on or interested in semantic matching, to share latest research results, express opinions on the related issues, and discuss future directions.
c/c++成长之路必备工具 @何_登成 太心水Google的各种工具了，试用他家的ThreadSanitizer O网页链接 好用！ThreadSanitizer/AddressSanitizer/MemorySanitizer，外加gperftools，工具覆盖了C/C++最难的两个领域：内存管理和多线程。招聘最好的员工，辅以最好的工具，其结果必然是产出最牛逼的系统。