Speaker: Kyunghyun Cho, University of Montreal
Location: Warren Weaver Hall 1302
Date: February 25, 2015, 11:30 a.m.
Host: Subhash Khot
Deep learning has become increasingly popular due to its success in challenging machine learning tasks such as large-scale object recognition, speech recognition and machine translation. In deep learning, we aim to build general computational models for data and tasks that exhibit rich, complex underlying structures, such as vision, speech and human languages. In this talk, I will give a brief overview of the principles behind deep learning and present in detail my latest research on neural machine translation and image/video description generation. I will conclude the talk by discussing the future of deep learning research, which I believe will let us gain insight into highly complicated natural phenomena.
Kyunghyun Cho is a postdoctoral researcher at the University of Montreal (Canada) under the supervision of Prof. Yoshua Bengio since early 2014. He has received B.Sc. in computer science from KAIST (Korea) in 2009. He continued his study at Aalto University (Finland) and received Ph.D. degree in 2014. His main research interest includes neural networks, generative models and their applications, especially, to language understanding.
Refreshments will be offered starting 15 minutes prior to the scheduled start of the talk.