IBM Cognitive Computing - An NLP Renaissance!
Speaker: Salim Roukos
Senior Manager of Multi-Lingual NLP and CTO for Translation Technologies at BM T. J. Watson Research Center
Date: Oct. 26, 2014
Electronically available multi-modal data (primarily text and meta-data) is unprecedented in terms of its volume, variety, velocity, (and veracity). The increased interest and investment in cognitive computing for building systems and solutions that enable and support richer human-machine interactions presents a unique opportunity for novel statistical models for natural language processing.
In this talk, I will describe a journey at IBM during the past three decades in developing novel statistical models for NLP covering statistical parsing, machine translation, and question-answering systems. Along with a discussion of some of the recent successes, I will discuss some difficult challenges that need to be addressed to achieve more effective cognitive systems and applications.
Salim Roukos is Senior Manager of Multi-Lingual NLP and CTO for Translation Technologies at IBM T. J. Watson Research Center. Dr. Roukos received his B.E. from the American University of Beirut, in 1976, his M.Sc. and Ph.D. from the University of Florida, in 1978 and 1980, respectively. He joined Bolt Beranek and Newman from 1980 through 1989, where he was a Senior Scientist in charge of projects in speech compression, time scale modification, speaker identification, word spotting, and spoken language understanding. He was an Adjunct Professor at Boston University in 1988 before joining IBM in 1989. Dr. Roukos has served as Chair of the IEEE Digital Signal Processing Committee in 1988.
Salim Roukos currently leads a group at IBM T.J. Watson research Center that focuses on various problems using machine learning techniques for natural language processing. The group pioneered many of the statistical methods for NLP from statistical parsing, to natural language understanding, to statistical machine translation and machine translation evaluation metrics (BLEU metric). Roukos has over a 150 publications in the speech and language areas and over two dozen patents. Roukos was the lead of the group which introduced the first commercial statistical language understanding system for conversational telephony systems (IBM ViaVoice Telephony) in 2000 and the first statistical machine translation product for Arabic-English translation in 2003. He has recently lead the effort to create IBM's offering of IBM Real-Time Translation Services (RTTS) a platform for enabling real-time translation applications such as multilingual chat and on-demand document translation.
Learning from Rational* Behavior
Speaker: Thorsten Joachims
Professor, the Department of Computer Science and the Department of Information Science, Cornell University
Date: Oct. 27, 2014
The ability to learn from user interactions can give systems access to unprecedented amounts of knowledge. This is evident in search engines, recommender systems, and electronic commerce, and it can be the key to solving other knowledge intensive tasks. However, extracting the knowledge conveyed by user interactions is less straightforward than standard machine learning, since it requires learning systems that explicitly account for human decision making, human motivation, and human abilities.
In this talk, I argue that the design space of such interactive learning systems encompasses not only the machine learning algorithm itself, but also the design of the interaction under an appropriate model of user behavior. To this effect, the talk explores how integrating microeconomic models of human behavior into the learning process leads to new interaction models and their associated learning algorithms, leading to systems that have provable guarantees and that perform robustly in practice.
* Restrictions apply. Some modeling required.
Thorsten Joachims is a Professor in the Department of Computer Science and the Department of Information Science at Cornell University. His research interests center on a synthesis of theory and system building in machine learning, with applications in information access, language technology, and recommendation. His past research focused on support vector machines, text classification, structured output prediction, convex optimization, learning to rank, learning with preferences, and learning from implicit feedback. In 2001, he finished his dissertation advised by Prof. Katharina Morik at the University of Dortmund. From there he also received his Diploma in Computer Science in 1997. Between 2000 and 2001 he worked as a PostDoc at the GMD Institute for Autonomous Intelligent Systems. From 1994 to 1996 he was a visiting scholar with Prof. Tom Mitchell at Carnegie Mellon University.