Representation-based Transfer Learning and Some Advances

Abstract

Current machine learning model is specific for the trained situation and performs poorly to others. What’s worse, the model adapted to a new task requires much corresponding labeled data. In the real world, it is hard to obtain sufficient labeled data for each task. Transfer learning boosts this learning processing through transferring previous knowledge to the new case. On the way to artificial intelligence, transfer learning is wanted. In the presentation, I will introduce representation-based transfer learning and some advances.

Date
Event
Lab Seminar
Location
Room 311, Innovation Center, UESTC
Avatar
Wei Han
Ph.D. Candidate

My research interests include interpretable machine learning, transfer Learning, continual / lifelong learning, few-shot learning and adversarial learning.