Dream To Control: Learning Behaviors By Latent Imagination

Abstract

Learned world models summarize an agent's experience to facilitate learning complex behaviors. While learning world models from high-dimensional sensory inputs is becoming feasible through deep learning, there are many potential ways for deriving behaviors from them. We present Dreamer, a reinforcement learning agent that solves long-horizon tasks from images purely by latent imagination. We efficiently learn behaviors by propagating analytic gradients of learned state values back through trajectories imagined in the compact state space of a learned world model. On 20 challenging visual control tasks, Dreamer exceeds existing approaches in data-efficiency, computation time, and final performance.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyhafner2019dream

Related papers

• Dream To Explore: Adaptive Simulations For Autonomous Systems (2021)0.00
• Curling The Dream: Contrastive Representations For World Modeling In Reinforcement Learning (2024)0.00
• Imagine-then-plan: Agent Learning From Adaptive Lookahead With World Models (2026)0.00
• Bridging Imagination And Reality For Model-based Deep Reinforcement Learning (2020)0.00
• Towards Biologically Plausible Dreaming And Planning In Recurrent Spiking Networks (2022)0.00
• Enhancing Policy Learning With World-action Model (2026)0.00
• Do Agents Dream Of Electric Sheep?: Improving Generalization In Reinforcement Learning Through Generative Learning (2024)0.00
• Dreaming Of Many Worlds: Learning Contextual World Models Aids Zero-shot Generalization (2024)2.83