- TL;DR: We present Dreamer, an agent that learns long-horizon behaviors purely by latent imagination using analytic value gradients.
- 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.
- Keywords: world model, latent dynamics, imagination, planning by backprop, policy optimization, planning, reinforcement learning, control, representations, latent variable model, visual control, value function
- Code: https://danijar.com/dreamer
- Original Pdf: pdf