2017-11-20 | 周熠:From First-Order Logic to Assertional Logic
题目
From First-Order Logic to Assertional Logic(PDF
摘要

First-order logic has been widely considered as the most important foundation for mathematical foundation. However, we argue that it is not a good foundation for knowledge representation, reasoning and learning as it has several critical drawback for this purpose.

Instead, we propose assertional logic, a new foundation for AI that targets on elegant, expressive and extensible knowledge representation.

报告人简介
周熠,现任澳大利亚西悉尼大学高级讲师,天津大学计算机科学与技术学院兼职教授。2001年、2006年分获中国科技大学学士和博士学位。在人工智 能、特别是知识表示与推理领域做出了重要贡献。是一阶回答集程序设计的奠基人及主要推动人之一,提出了首个刻画遗忘的公理系统等。在人工智能顶级期刊Artificial Intelligence上发表6篇长文。长期担任人工智能顶级会议程序设计委员会委员,包括IJCAI、AAAI、KR等等。
2017-10-23 | Jyrki Nummenmaa:Towards utilisation of grammatical information in question answering
题目
Towards utilisation of grammatical information in question answering
摘要

Text analysis is a fundamental task with many applications such as information retrieval, analysis of recommendations, question answering, and various task related with AI. The bag-of-words and k-gram based methods have, as initial basic approaches, have been followed by the use of machine learning techniques such as deep learning. However, grammar is what creates conceptual content out of words. Therefore, we suggest a grammatical analysis of text as a basis for analysing texts. We present an example case where we apply the grammatical approach in a particular question answering task. We get results comparable to the use of neural networks, however the results are also explainable.

In addition, the talk will contain an introduction to information retrieval, text analysis, and web mining research at the University of Tampere.

报告人简介
Mr. Jyrki Nummenmaa is a full professor at the School of Information Sciences of University of Tampere, Finland and the head of Research Center for Information and Systems (CIS) at the University of Tampere. Prof. Nummenmaa has done research on algorithms, databases, software development, business intelligence, data mining, open data, and Big Data, and, most recently, in text analysis. He has extensive administrational experience and he has practical experience of working 3,5 years in software companies in Tampere area. He has visited the University of Edinburgh for one year while doing his PhD research and lately several times universities in China, in 2004 University of Chile for 2 months, and in 2017 IT Faculty of Chalmers and University of Gothenburg for 2 months. He has over 70 peer-reviewed scientific publications in scientific journals and conferences.
2017-10-10 | Yu Su:On Building Natural Language Interface to Data and Services
题目
On Building Natural Language Interface to Data and Services(PDF
摘要
By mapping natural languages to the formal languages used by computers, natural language interface holds the promise for serving as a unified interface to seamlessly access and interact with a great variety of data and services. For decades people have long been envisaging generic natural language interfaces that grant all users, regardless of their proficiency in programming, equal access to the digital world. With the recent advances in deep learning and knowledge base, it is widely believed that we are better equipped than ever to tackle some most challenging problems in this area, and practical, well-functioning natural language interfaces for a wide range of domains are about to be born. In this talk, I will present our recent explorations in natural language interface. The talk will start with discussion on a holistic design of practical natural language interfaces, covering important topics like semantic parsing, scalability, interpretability and interactivity. I will then specifically address one challenging question: Given a new domain with **zero** training data and user, how to build a natural language interface from scratch? We discuss two complementary solutions, one using crowdsourcing to collect training data at a low cost, and the other using existing data from other domains via neural transfer learning. Finally, I will briefly describe our studies towards making natural language interfaces more robust and usable, such as answer triggering, i.e., checking whether a question is answerable, and post-inspection, i.e., checking whether the parsing result is sensible before execution. I will end the talk with a few exciting future directions including cross-domain natural language interface and application to important vertical domains like healthcare and finance.
报告人简介
Yu Su is a final-year Ph.D. candidate in the Department of Computer Science at University of California, Santa Barbara. He works with Xifeng Yan on a variety of topics at the intersection of data mining, natural language processing, and machine learning. Before coming to UCSB, he got his bachelor degree from the Department of Computer Science and Technology at Tsinghua University in 2012. His recent research interests include natural language interface, knowledge base construction, and general text mining, with a focus on understanding the interplay of natural and formal languages to increase the accessibility of data (e.g., knowledge bases, web tables, and relational databases) and services (e.g., web APIs). He has published over 10 papers on question answering/semantic parsing, graph mining and querying, and crowdsourcing at major conferences including SIGKDD, WWW, EMNLP, and CIKM. He has served on the program committee or as reviewer for many top conferences and journals including WWW, ACL, EMNLP, and TKDE. He has interned at Microsoft Research Redmond, IBM T.J. Watson Research Center, and U.S. Army Research Laboratory.