Adaptive Reinforcement Learning through Evolving Self-Modifying Neural Networks
This addresses the challenge of real-time adaptation in robotics for locomotion, though it is incremental as it builds on prior neuromodulated plasticity methods.
The paper tackled the problem of enabling online adaptation in reinforcement learning for complex meta-learning tasks, specifically in a quadruped domain where limbs can become unusable, and found that agents with self-modifying plastic networks outperformed gradient-based methods and trained faster.
The adaptive learning capabilities seen in biological neural networks are largely a product of the self-modifying behavior emerging from online plastic changes in synaptic connectivity. Current methods in Reinforcement Learning (RL) only adjust to new interactions after reflection over a specified time interval, preventing the emergence of online adaptivity. Recent work addressing this by endowing artificial neural networks with neuromodulated plasticity have been shown to improve performance on simple RL tasks trained using backpropagation, but have yet to scale up to larger problems. Here we study the problem of meta-learning in a challenging quadruped domain, where each leg of the quadruped has a chance of becoming unusable, requiring the agent to adapt by continuing locomotion with the remaining limbs. Results demonstrate that agents evolved using self-modifying plastic networks are more capable of adapting to complex meta-learning learning tasks, even outperforming the same network updated using gradient-based algorithms while taking less time to train.