Do Agents Dream Of Electric Sheep?: Improving Generalization In Reinforcement Learning Through Generative Learning

Abstract

The Overfitted Brain hypothesis suggests dreams happen to allow generalization in the human brain. Here, we ask if the same is true for reinforcement learning agents as well. Given limited experience in a real environment, we use imagination-based reinforcement learning to train a policy on dream-like episodes, where non-imaginative, predicted trajectories are modified through generative augmentations. Experiments on four ProcGen environments show that, compared to classic imagination and offline training on collected experience, our method can reach a higher level of generalization when dealing with sparsely rewarded environments.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyfranceschelli2024do

Related papers

• Illuminating Generalization In Deep Reinforcement Learning Through Procedural Level Generation (2018)0.00
• Towards Biologically Plausible Dreaming And Planning In Recurrent Spiking Networks (2022)0.00
• Task Aware Dreamer For Task Generalization In Reinforcement Learning (2023)0.00
• Dream To Control: Learning Behaviors By Latent Imagination (2019)0.00
• Genai-based Multi-agent Reinforcement Learning Towards Distributed Agent Intelligence: A Generative-rl Agent Perspective (2025)0.00
• Improving Generalization To New Environments And Removing Catastrophic Forgetting In Reinforcement Learning By Using An Eco-system Of Agents (2022)0.00
• Learning As Reinforcement: Applying Principles Of Neuroscience For More General Reinforcement Learning Agents (2020)0.00
• Learning Of Generalizable And Interpretable Knowledge In Grid-based Reinforcement Learning Environments (2023)3.58