Graceful task adaptation with a bi-hemispheric RL agent
Not on the map. Why?
Abstract (arXiv)
In humans, responsibility for performing a task gradually shifts from the right hemisphere to the left. The Novelty-Routine Hypothesis (NRH) states that the right and left hemispheres are used to perform novel and routine tasks respectively, enabling us to learn a diverse range of novel tasks while performing the task capably. Drawing on the NRH, we develop a reinforcement learning agent with specialised hemispheres that can exploit generalist knowledge from the right-hemisphere to avoid poor initial performance on novel tasks. In addition, we find that this design has minimal impact on its ability to learn novel tasks. We conclude by identifying improvements to our agent and exploring potential expansion to the continual learning setting.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 12 of its 30 references that are in the map. First posted on arXiv 16 Jul 2024.
Builds on
12 of 30 references.
References · 12
- 2017Proximal Policy Optimization Algorithmscited by 14,441
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,218
- 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,313
- 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
- 2016RL2: Fast Reinforcement Learning via Slow Reinforcement Learningcited by 490
- 2016Learning to reinforcement learncited by 369
- 2020Towards Continual Reinforcement Learning: A Review and Perspectivescited by 268
- 2021A Survey of Zero-shot Generalisation in Deep Reinforcement Learningcited by 164
- 2021Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPscited by 115
- 2021Continual World: A Robotic Benchmark For Continual Reinforcement Learningcited by 112
- 2019Meta-learning of Sequential Strategiescited by 68
Built on by
Not cited yet.