2 papers
cs.LG2026
Memory Merge DQN: Sensitivity Weighted Target Updates for Stable Value Learning
Adrian Ly, Richard Dazeley, Peter Vamplew +2
Deep Q-networks use target networks to stabilise bootstrapped value learning, but the standard hard copy update also introduces a tradeoff. Holding the target network fixed, improv…
cs.LG2026
Ensemble Elastic DQN: A Step Dependent Ensemble Approach for Reducing Overestimation in Deep Value-Based Reinforcement Learning
Adrian Ly, Richard Dazeley, Peter Vamplew +2
Deep Q-Networks (DQN) can suffer from overestimation bias because bootstrapped targets use a maximisation operation over noisy value estimates. Ensemble-based methods and multi-ste…