Dr. Lili Mou is an Assistant Professor at the Department of Computing Science, University of Alberta. Lili received his BS and PhD degrees in 2012 and 2017, respectively, from School of EECS, Peking University. After that, he worked as a postdoctoral fellow at the University of Waterloo and a research scientist at Adeptmind (a startup in Toronto, Canada). His research interests include deep learning applied to natural language processing as well as programming language processing. He has publications at top conferences and journals, including AAAI, ACL, CIKM, COLING, EMNLP, ICASSP, ICLR, ICML, IJCAI, INTERSPEECH, NAACL-HLT, and TACL (in alphabetic order). He is supported by AltaML, Amii Fellow Program, and Canadian CIFAR AI Chair Program.
My research mission is to build an intelligent system that can understand and interact with humans via natural language, involving both text understanding and generation. Towards this long-term goal, I am focusing on fundamental problems in machine learning (especially, deep learning) methods applied to natural language processing, including feature extraction in the discrete input space, weakly supervised learning in the discrete latent space, and sentence synthesis in the discrete output space. My work has been successfully applied to various NLP tasks, including information extraction, semantic parsing, syntactic parsing, text generation, and many others.
CMPUT 651: Topics in Artificial Intelligence (Deep Learning for Natural Language Processing)
This course introduces deep learning (DL) techniques for natural language processing (NLP). Contrary to other DL4NLP courses, we would have a whirlwind tour of all neural architectures (e.g., CNNs, RNNs, attention) in a few lectures. Then, we would make significant efforts in learning structured prediction using Bayesian and Markov networks, with applications of sequential labeling, syntactic parsing, and sentence generation. In this process, we will also see how such traditional methods can be combined with and improve a plain neural network.
Learning is essential for many real-world tasks, including recognition, diagnosis, forecasting and data-mining. This course covers a variety of learning scenarios (supervised, unsupervised and partially supervised), as well as foundational methods for regression, classification, dimensionality reduction and modeling. Techniques such as kernels, optimization and probabilistic graphical models will typically be introduced. It will also provide the formal foundations for understanding when learning is possible and practical. Prerequisites: one of CMPUT 340 or 418; one of STAT 141, 151, 235 or 265 or SCI 151; or consent of the instructor.Fall Term 2020