A stochastic gating mechanism that induces controlled partial observability during training by sampling a spatial binary mask from a Beta--Bernoulli random field and applying it to the raw observation is proposed.
Abstract
Model-based reinforcement learning (MBRL) has enabled sample-efficient learning of continuous control policies from high-dimensional observations through learned latent dynamics models. These models are typically trained with observation reconstruction objectives, which provide strong supervision but can bias representations toward visual fidelity leading to an undertrained prior and unreliable long-horizon predictions. We propose a stochastic gating mechanism that induces controlled partial observability during training by sampling a spatial binary mask from a Beta--Bernoulli random field and applying it to the raw observation. The gate probability is resampled at each time step, exposing the encoder to conditions ranging from near-total occlusion to near-full visibility. Integration of our method requires minimal architectural changes from existing dynamics models: the gated observation and the binary mask are channel-wise concatenated and passed to the encoder as a single augmented input, with no modifications to the broader training pipeline. We evaluate the proposed approach on a set of pixel-based continuous control tasks selected from the DeepMind Control Suite. Using model predictive control for evaluation, which relies solely on open-loop model rollouts, we show that stochastic gating improves planning performance and data efficiency across tasks. Further analysis demonstrates that performance is robust to the spatial scale of the mask, and a balanced observability during training outperforms configurations biased toward either sparser or denser observations.
Successor-feature-based zero-shot reinforcement learning methods pre-train a task encoder and successor-feature predictor on offline, reward-free data, then recover a policy at test time by solving a closed-form linear regression over a small labeled dataset. We identify an underexplored limitation of this approach: wh...
This work revisits Unitary recurrent networks (uRNNs) and proposes three different versions of uRNNs as drop-in replacements for recurrent PPO architectures, and demonstrates that the simple recurrence and the added degree of freedom from the phase of the complex representations enable significant gains over baselines...
Sathya Kamesh Bhethanabhotla, E. Gavves, André Biedenkapp· arXiv.org· 0 citations
Learning from trajectory demonstrations offers a route to active-inference control of complex systems whose dynamics are difficult to model explicitly. We introduce generative active-inference control (GenAIF), in which one generative trajectory model learns from demonstrations and measured action interventions to supp...
Yu-Lin Li, Mohsen A. Jafari, Andrea Matta· 0 citations
Model-based reinforcement learning (MBRL) achieves strong sample efficiency by planning in learned latent dynamics but degrades substantially under unseen visual distractions such as background, lighting, or camera changes. Unlike model-free RL, where encoder perturbations affect only single-step predictions, MBRL suff...
Mingyu Park, Samyeul Noh, Hyun Myung et al.· 0 citations
Recent autonomous driving systems have increasingly shifted from modular pipelines to end-to-end learning paradigms. However, purely end-to-end approaches are prone to shortcut learning or representation collapse, often requiring auxiliary perception or prediction tasks to improve generalization. These auxiliary object...
Shuai Liu, Si-Heng Ren, Xiao-Yao Zhu et al.· 0 citations
This work introduces density-guided bootstrap boundaries, a trajectory-dependent mechanism that adapts the depth of imagined returns using a score-informed relative density proxy and clarifies the theoretical assumptions distinguishing the authors' expansion proxy from exact path likelihoods.
Shashank Gupta, Pilhwa Lee· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.