DeepMind Control

DeepMind Control is a suite of continuous control tasks designed to evaluate reinforcement learning algorithms, focusing on state-based environments.

Papers using DeepMind Control (11)

Skill Learning via Policy Diversity Yields Identifiable Representations for Reinforcement Learning2025 · 3 citesReflex: Reinforcement Learning with Reflection Symmetry Exploitation in State-Based Continuous Control2026Dreaming Smoothly and Sample Efficiently with Gradient Penalized Latent Dynamics2026GIRL: Generative Imagination Reinforcement Learning via Information-Theoretic Hallucination Control2026Unifying Model-Free Efficiency and Model-Based Representations via Latent Dynamics2026WIMLE: Uncertainty-Aware World Models with IMLE for Sample-Efficient Continuous Control2026Revisiting Bisimulation Metric for Robust Representations in Reinforcement Learning2025Scaling CrossQ with Weight Normalization2025Certifying Stability of Reinforcement Learning Policies using Generalized Lyapunov Functions2025Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control2024 · 1 citesState Chrono Representation for Enhancing Generalization in Reinforcement Learning2024 · 1 cites