Annealing the critic's expectile parameter from near-max to mean (Annealed Q-learning) improves sample efficiency and final scores of TD3 and SAC on continuous control benchmarks.
S., Courville, A., and Bellemare, M
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Gradual Transition from Bellman Optimality Operator to Bellman Operator in Online Reinforcement Learning
Annealing the critic's expectile parameter from near-max to mean (Annealed Q-learning) improves sample efficiency and final scores of TD3 and SAC on continuous control benchmarks.