| Eugene Charniak |
| Fri 13 Nov 2015, 11:00 - 12:30 |
| Informatics Forum (IF-4.31/4.33) |
If you have a question about this talk, please contact: Diana Dalla Costa (ddallac)
Abstract:
We develop a model of recipe understanding as a noisy channel model --- i.e., treat the problem as one of translation from English to a robot cooking language. As we do not have such a language we use CURD, a simplified recipe description language developed by Tassi and Smith. We present several such models and in particular show the importance of the cooking prior in the performance of the system. However, as little of this works at the time of this writing, the title and contents of this talk are subject to change without prior notice.
Bio:
Eugene Charniak is University Professor of Computer Science at Brown University. He received his AB (with honors) degree in Physics from University of Chicago, and his MS in Electrical Engineering and PhD in Computer Science from MIT. He is a fellow of the Association for the Advancement of Artificial Intelligence (AAAI) and the Association for Computational Linguistics (ACL). He has received a classic paper award from AAAI, and the lifetime achievement award from ACL.