2 papers
cs.NE2025
Online Continual Learning via Spiking Neural Networks with Sleep Enhanced Latent Replay
Erliang Lin, Wenbin Luo, Wei Jia +2
Edge computing scenarios necessitate the development of hardware-efficient online continual learning algorithms to be adaptive to dynamic environment. However, existing algorithms…
cs.LG2025
A Variance-Reduced Cubic-Regularized Newton for Policy Optimization
Cheng Sun, Zhen Zhang, Shaofu Yang
In this paper, we study a second-order approach to policy optimization in reinforcement learning. Existing second-order methods often suffer from suboptimal sample complexity or re…