17 citations · 35 across the 13 of their papers we have counts for
13 papers
AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projection
Saleh Momeni, Changnan Xiao, Bing Liu
This paper studies the problem of class-incremental learning (CIL), a core setting within continual learning where a model learns a sequence of tasks, each containing a distinct se…
A Theory for Length Generalization in Learning to Reason
Changnan Xiao, Bing Liu
Length generalization (LG) is a challenging problem in learning to reason. It refers to the phenomenon that when trained on reasoning problems of smaller lengths or sizes, the resu…
Conditions for Length Generalization in Learning Reasoning Skills
Changnan Xiao, Bing Liu
Reasoning is a fundamental capability of AI agents. Recently, large language models (LLMs) have shown remarkable abilities to perform reasoning tasks. However, numerous evaluations…
Learnability and Algorithm for Continual Learning
Gyuhak Kim, Changnan Xiao, Tatsuya Konishi +1
This paper studies the challenging continual learning (CL) setting of Class Incremental Learning (CIL). CIL learns a sequence of tasks consisting of disjoint sets of concepts or cl…
Open-World Continual Learning: Unifying Novelty Detection and Continual Learning
Gyuhak Kim, Changnan Xiao, Tatsuya Konishi +2
As AI agents are increasingly used in the real open world with unknowns or novelties, they need the ability to (1) recognize objects that (a) they have learned before and (b) detec…
Mastering Strategy Card Game (Legends of Code and Magic) via End-to-End Policy and Optimistic Smooth Fictitious Play
Wei Xi, Yongxin Zhang, Changnan Xiao +5
Deep Reinforcement Learning combined with Fictitious Play shows impressive results on many benchmark games, most of which are, however, single-stage. In contrast, real-world decisi…
Mastering Strategy Card Game (Hearthstone) with Improved Techniques
Changnan Xiao, Yongxin Zhang, Xuefeng Huang +3
Strategy card game is a well-known genre that is demanding on the intelligent game-play and can be an ideal test-bench for AI. Previous work combines an end-to-end policy function…
A Theoretical Study on Solving Continual Learning
Gyuhak Kim, Changnan Xiao, Tatsuya Konishi +2
Continual learning (CL) learns a sequence of tasks incrementally. There are two popular CL settings, class incremental learning (CIL) and task incremental learning (TIL). A major c…
Generalized Data Distribution Iteration
Jiajun Fan, Changnan Xiao
To obtain higher sample efficiency and superior final performance simultaneously has been one of the major challenges for deep reinforcement learning (DRL). Previous work could han…
Continual Learning Based on OOD Detection and Task Masking
Gyuhak Kim, Sepideh Esmaeilpour, Changnan Xiao +1
Existing continual learning techniques focus on either task incremental learning (TIL) or class incremental learning (CIL) problem, but not both. CIL and TIL differ mainly in that…
GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning
Jiajun Fan, Changnan Xiao, Yue Huang
Deep Q Network (DQN) firstly kicked the door of deep reinforcement learning (DRL) via combining deep learning (DL) with reinforcement learning (RL), which has noticed that the dist…
An Entropy Regularization Free Mechanism for Policy-based Reinforcement Learning
Changnan Xiao, Haosen Shi, Jiajun Fan +1
Policy-based reinforcement learning methods suffer from the policy collapse problem. We find valued-based reinforcement learning methods with ε-greedy mechanism are capable of enjo…
CASA: Bridging the Gap between Policy Improvement and Policy Evaluation with Conflict Averse Policy Iteration
Changnan Xiao, Haosen Shi, Jiajun Fan +2
We study the problem of model-free reinforcement learning, which is often solved following the principle of Generalized Policy Iteration (GPI). While GPI is typically an interplay…