Mind Dreamer: Untethering Imagination via Active Latent Intervention on Latent Manifolds
Shaojun Xu, Xiaoling Zhou, Yihan Lin, Yapeng Meng, Xinglong Ji, Luping Shi, Rong Zhao

TL;DR
Mind Dreamer introduces Active Latent Intervention to enhance model-based RL by synthesizing plausible non-continuous latent states, improving exploration and sample efficiency in complex environments.
Contribution
It proposes a novel framework that transcends Markovian constraints using adversarial latent jumps and new value functions, advancing latent imagination in RL.
Findings
Achieves 1.67× speedup over DreamerV3 on DeepMind Control Suite.
Reaches 8.8× speedup in sparse-reward tasks.
Theoretically establishes quadratic discount for uncertainty propagation.
Abstract
Model-Based Reinforcement Learning (MBRL) leverages latent imagination for sample efficiency, yet remains constrained by Historical Tethering: imagination is typically initialized from observed states. This creates a learning asymmetry, where the world model's manifold discovery outpaces the policy's sparse-reward optimization. We propose Mind Dreamer (MD), a framework that operationalizes Active Latent Intervention (ALI) to transcend Markovian continuity. MD reformulates discovery as the minimization of a global Relay Manifold Expected Free Energy (R-EFE); by sampling initial states from a learned generator rather than the historical buffer, MD utilizes an adversarial generator to synthesize non-continuous latent jumps to epistemic blind spots that are physically plausible yet cognitively challenging. To resolve the credit assignment paradox across these…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
