mlsanityData to 14 Oct 2026

Learning High-Risk High-Precision Motion Control

arXiv 2609.34851 ↗First posted 28 Sep 2026cs.LG cs.ROcited by 0

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

Deep reinforcement learning (DRL) algorithms for movement control are typically evaluated and benchmarked on sequential decision tasks where imprecise actions may be corrected with later actions, thus allowing high returns with noisy actions. In contrast, we focus on an under-researched class of high-risk, high-precision motion control problems where actions carry irreversible outcomes, driving sharp peaks and ridges to plague the state-action reward landscape. Using computational pool as a representative example of such problems, we propose and evaluate State-Conditioned Shooting (SCOOT), a novel DRL algorithm that builds on advantage-weighted regression (AWR) with three key modifications: 1) Performing policy optimization only using elite samples, allowing the policy to better latch on to the rare high-reward action samples; 2) Utilizing a mixture-of-experts (MoE) policy, to allow switching between reward landscape modes depending on the state; 3) Adding a distance regularization term and a learning curriculum to encourage exploring diverse strategies before adapting to the most advantageous samples. We showcase our features' performance in learning physically-based billiard shots demonstrating high action precision and discovering multiple shot strategies for a given ball configuration.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 15 of its 44 references that are in the map. First posted on arXiv 28 Sep 2026.

Builds on

15 of 44 references.

References · 15

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  4. 2018DeepMind Control Suitecited by 956
  5. 2019Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learningcited by 782
  6. 2019On the Variance of the Adaptive Learning Rate and Beyondcited by 772
  7. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 407
  8. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 258
  9. 2019MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policiescited by 108
  10. 2019Reward-Conditioned Policiescited by 102
  11. 2020Sample-efficient Cross-Entropy Method for Real-time Planningcited by 95
  12. 2015Learning Visual Predictive Models of Physics for Playing Billiardscited by 58
3 more
  1. 2018PPO-CMA: Proximal Policy Optimization with Covariance Matrix Adaptationcited by 14
  2. 2022A Survey on Reinforcement Learning Methods in Character Animationcited by 12
  3. 2019Visualizing Movement Control Optimization Landscapescited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2015Continuous control with deep reinforcement learning
  2. 2025Learning to Ball: Composing Policies for Long-Horizon Basketball Moves
  3. 2026Enhancing Control Policy Smoothness by Aligning Actions with Predictions from Preceding States
  4. 2025Dynamic Legged Ball Manipulation on Rugged Terrains with Hierarchical Reinforcement Learning
  5. 2024Absolute State-wise Constrained Policy Optimization: High-Probability State-wise Constraints Satisfaction
  6. 2025Control-Optimized Deep Reinforcement Learning for Artificially Intelligent Autonomous Systems
  7. 2026Stochastic Multiple Shooting Trajectory Optimization via Sequential Local Policy Evaluation ↗
  8. 2022GoalsEye: Learning High Speed Precision Table Tennis on a Physical Robot