Zijian Gao

class-incremental learning (2) catastrophic forgetting (2) offline reinforcement learning (1) policy optimization (1) knowledge distillation (1) loss landscape (1) out-of-distribution generalization (1) markov decision process (1) model-based reinforcement learning (1) pre-trained model (1) vision-language model (1) cross-entropy loss (1) parameter-efficient adaptation (1) online continual learning (1) exponential moving average (1) knowledge retention (1) knowledge memorization (1) recency bia (1) logit alignment (1) pessimistic markov decision process (1)