Hee bin Yoo

reinforcement learning (1) sample efficiency (1) metric learning (1) policy optimization (1) embedding learning (1) feature embedding (1) machine unlearning (1) neural collapse (1) decision boundary (1) sparse reward (1) classifier head (1) selective unlearning (1) neural network (1) reward transition (1) potential-based reward (1) weight drift (1) classifier initialization (1)