Disentangling Transfer in Continual Reinforcement Learning
–Neural Information Processing Systems
We adopt SAC as the underlying RL algorithm and Continual World as a suite of continuous control tasks. We systematically study how different components of SAC (the actor and the critic, exploration, and data) affect transfer efficacy, and we provide recommendations regarding various modeling options.
Neural Information Processing Systems
Oct-3-2025, 04:03:56 GMT
- Country:
- Asia > Japan
- Honshū > Kansai > Osaka Prefecture > Osaka (0.04)
- Europe
- Germany > Bavaria
- Upper Bavaria > Munich (0.04)
- Poland
- Lesser Poland Province > Kraków (0.04)
- Masovia Province > Warsaw (0.04)
- United Kingdom > England
- Greater London > London (0.04)
- Germany > Bavaria
- North America
- Canada > British Columbia
- Vancouver (0.04)
- United States
- California > Los Angeles County
- Long Beach (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Nevada (0.04)
- Utah > Salt Lake County
- Salt Lake City (0.04)
- California > Los Angeles County
- Canada > British Columbia
- Asia > Japan
- Genre:
- Overview (0.46)
- Research Report (0.46)
- Technology: