Growing Action Spaces
cs.ox.ac.uk · 7,225 words · saved by 1 readers
N/A
Growing Action Spaces Gregory Farquhar 1 2 Laura Gustafson 3 Zeming Lin 3 Shimon Whiteson 1 Nicolas Usunier 3 Gabriel Synnaeve 3 Abstract neither the agent nor experimenter has such unimpeded con- trol over the environment. In this work, we instead make In complex tasks, such as those with large com- use of curricula that are internal to the…
saved by
related reading
- 2312.10812arxiv.org
- Extended Data Fig. 5: Training progress. | Naturenature.com
- Multi-task curriculum learning in a complex, visual,hard-exploration domain: Minecraftarxiv.org
- A (Long) Peek into Reinforcement Learning | Lil'Loglilianweng.github.io
- Part 1: Key Concepts in RL - Spinning Up documentationspinningup.openai.com
- Learning Beyond Gradientstrinkle23897.github.io
- The Promise of Hierarchical Reinforcement Learningthegradient.pub
- Exploration for the Efficient Deployment of Reinforcement Learning Agentsopenreview.net
- Teacher Algorithms for Deep RL Agents that Generalize in Procedurally Generated Environments – Developmental Systems, a Blog of the Flowers Labdevelopmentalsystems.org
- How DeepMind's Generally Capable Agents Were Trained — LessWronglesswrong.com
- Key Papers in Deep RL - Spinning Up documentationspinningup.openai.com
- Ch. 21 - Imitation Learningunderactuated.mit.edu