Skip to main content
Search

# Model-Based Reinforcement Learning

4 related articles

Subscribe to updates
  1. arXiv

    ScanDreamer gives quadrotors an online map to imagine with: 96.3% success and 3.6% collisions in unseen simulated layouts

    For quadrotor navigation through unknown clutter, ScanDreamer builds an online 2D scan-memory map from depth images and vehicle pose, and conditions its world model, actor and critic on that map, so imagined rollouts advance the pose over a frozen map and read geodesic progress and wall-contact signals from mapped geometry. Across 1560 test episodes in NVIDIA Isaac Sim spanning unseen layouts and parameter ranges, three seeds average 96.3% success and 3.6% collisions, against 59.6% success and 28.8% collisions for AirDreamer at the same online-interaction budget.

    Nanyang Technological UniversityVinUniversity
    Yunyi Gao（First author）
    ScanDreamerQuadrotor NavigationWorld ModelOccupancy MappingModel-Based Reinforcement Learning
  2. arXiv

    Assistive agents in shared driving learn to guess intent first, improving torque alignment and collision rate in racing and urban driving

    Assistive agents in shared control must act on the human's intent while staying legible enough for the human to respond in real time. Dream2Assist adds intent inference to a recurrent state-space world model and trains the assistive policy through fictitious co-play, anticipating discrete human objectives without overriding human agency. Three human-subjects studies and a preliminary on-vehicle deployment span high-speed racing (N=30) and urban driving with richer intent structure (N=10), where intent conditioning yields measurable improvements in torque alignment and collision rate.

    Jonathan DeCastro（First author）
    Dream2AssistModel-Based Reinforcement LearningWorld ModelIntent InferenceShared Decision Making
  3. arXiv

    World models stop stepping one at a time: one forward pass yields H-step hypothetical rollouts for humanoid and quadruped control

    Model-based reinforcement learning now uses a chunk world model as a differentiable training-time evaluator, generating H-step hypothetical rollouts in a single forward pass and directly optimizing value expansion, evaluated on 19 high-dimensional humanoid tasks from HumanoidBench and deployed to real-world quadruped and bipedal walking.

    Shanghai Innovation InstituteShanghai Jiao Tong UniversityWestlake University
    Weixi Song（First author）
    Chunk World ModelModel-Based Reinforcement LearningHumanoid Locomotion
  4. arXiv

    VIGOR uses latent-space consistency to give model-based RL zero-shot generalization to unseen visual distractions, beating the second-best baseline by 3.4% on DMC and 43.6% on Robosuite

    The authors propose VIGOR, a framework combining asymmetric weak-to-strong augmentation, dynamics-level consistency via direct latent regression that enforces augmentation-invariant transition predictions, and encoder-level stabilization, achieving zero-shot generalization to unseen visual distractions while retaining the sample efficiency of its model-based RL backbone; it outperforms state-of-the-art model-free and model-based baselines on the DeepMind Control Suite and Robosuite, surpassing the second-best baseline by 3.4% and 43.6% respectively, and ablations show that replacing the default augmentation with alternatives from distinct perturbation families preserves strong generalization, indicating that latent-space consistency rather than the augmentation choice drives robustness.

    Mingyu Park（First author）
    VIGORLatent-Space ConsistencyZero-Shot Visual GeneralizationModel-Based Reinforcement Learning

You’re all caught up