Aidong Zhang

Title: Transfer Learning and Meta-Learning for Few-Shot Learning Applications

Abstract: Transfer learning has been proposed to re-use the trained model parameters in similar machine learning applications. However, transfer learning cannot be effectively applied from one domain to a different domain. Recently, meta-learning, which utilizes prior knowledge learned from related tasks and generalizes to new tasks of limited supervised experience, has shown to be an effective approach for few-shot learning probems. In this talk, I will discuss the advantage of meta-learning over traditional classification in prediction problems and use The Cancer Genome Atlas (TCGA) cancer patient data to demonstrate that meta-learning can outperform the conventional transfer learning in cancer prediction problems. I will also present a knowledge-guided meta-learning strategy which integrates biological knowledge with meta-learning for improved classification performance on few-shot learning problems.

Bio: Dr. Aidong Zhang is a William Wulf Faculty Fellow and Professor of Computer Science in the School of Engineering and Applied Sciences at University of Virginia (UVA). She also holds joint appointments with Department of Biomedical Engineering and Data Science Institute at University of Virginia. Her research interests include machine learning, data science, bioinformatics, and health informatics. Dr. Zhang is a fellow of ACM, AIMBE, and IEEE.

Ed Chi

Title: Efficient Neural Modeling for Large-Scale Real-World Recommendations

Abstract: Fundamental improvements in recommendation and ranking have been much harder to come by, when compared with recent progress on other long-standing AI problems such as computer vision, speech recognition, language models, and machine translation. Some reasons include: (1) large amounts of data making training difficult, yet having (2) noisy and sparse labels; (3) changing dynamics of context such as user preferences and items; and (4) low-latency requirement for a recommendation response. Finally, (5) the data are skewed long-tailed distributions (i.e. power law). That is, while there is a good amount of logged data, but for many users and items in the long tail, we actually suffer from extreme sparsity of labels.) These problems means we have to make very efficient use of data to provide the personalized recommendation experiences that users expect. In this talk, we will touch upon many recent advances in data-efficient neural modeling techniques for recommendations and their impact in real-world products. We will particularly focus on (a) using multi-task models with gated mixture of experts to jointly learn tasks with different label sparsity; and (b) employing recent Self-Supervised Learning techniques to robustly learn feature representations; and finish up with (c) some thoughts on the relationships between data efficiency and compressive sensing.

Bio: Ed H. Chi is a Distinguished Scientist at Google, leading several machine learning research teams focusing on neural modeling, reinforcement learning, dialog modeling, reliable/robust machine learning, and recommendation systems in the Google Brain team. His team has delivered significant improvements for YouTube, News, Ads, Google Play Store at Google with >420 product improvements since 2013. With 39 patents and >150 research articles, he is also known for research on user behavior in web and social media. Prior to Google, he was the Area Manager and a Principal Scientist at Palo Alto Research Center's Augmented Social Cognition Group, where he led the team in understanding how social systems help groups of people to remember, think and reason. Ed completed his three degrees (B.S., M.S., and Ph.D.) in 6.5 years from University of Minnesota. Recognized as an ACM Distinguished Scientist and elected into the CHI Academy, he recently received a 20-year Test of Time award for research in information visualization. He has been featured and quoted in the press, including the Economist, Time Magazine, LA Times, and the Associated Press. An avid swimmer, photographer and snowboarder in his spare time, he also has a blackbelt in Taekwondo.