Continuous Control

Follow this topic Ask about this topic

Momentum

3 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 34

Oct 1, 2026cs.LG

FERPO: Forward Entropy-Regularized Policy Optimization

Several state-of-the-art methods for online reinforcement learning in continuous control improve policies using action gradients of a learned critic. However, critics are typically trained to predict returns, and accurate value predictions do not necessarily yield accurate action derivatives, potentially leading to unreliable policy updates. We propose Forward Entropy-Regularized Policy Optimization (FERPO), an on-policy maximum entropy reinforcement learning algorithm that performs policy improvement using critic values without differentiating the critic with respect to actions. FERPO derives an optimal target action distribution from a policy-improvement objective regularized by entropy and Kullback-Leibler (KL) divergence. We then fit the actor to this target by minimizing a forward-KL objective, estimated using self-normalized importance sampling (SNIS) with actions drawn from the rollout policy. By limiting the target distribution's deviation from the rollout policy, the KL regularization helps keep these importance weights well behaved. In contrast to reverse-KL objectives, which can favor a subset of the target distribution's modes, the forward-KL objective encourages coverage of multiple high-value modes and thereby promotes exploration. Experiments and ablations on MuJoCo Playground and ManiSkill show competitive performance and sample-efficiency gains. Computational benchmarks also demonstrate faster actor updates than Relative Entropy Pathwise Policy Optimization (REPPO).

Sebastian Sanokowski, Alireza Sarmadi, Majid Khadiv

Applied and Theoretical Aspects of Robot Intelligence (ATARI) Lab, Munich Institute of Robotics and Machine Intelligence (MIRMI), Technical University of Munich

KL-Regularized RL Policy Optimization Maximum Entropy RL Continuous Control Policy Learning

Sep 17, 2026cs.RO

GR2PO: Group Relative Return Policy Optimization for Continuous Robot Control

Actor-critic architecture has been widely used in continuous robot control. However, they rely on learning a value network, introducing additional computational overhead during training. Moreover, policy learning may also be affected by the approximation error of value estimation. Critic-free group relative policy optimization methods provide a simpler training approach by removing the need for a critic. However, they fail to learn long-term action outcomes when directly applying immediate rewards to policy optimization in dense-reward environments. To address these problems, we propose Group Relative Return Policy Optimization (GR2PO), a critic-free reinforcement learning framework for continuous robot control. GR2PO estimates the discounted returns from the parallelly collected trajectories, performs group normalization at each rollout time index, and uses relative advantages and clipped targets to update the policy. To evaluate the effectiveness of the proposed framework, we instantiate it on robot control simulation environments and deploy the model to a real-world edge device. The results show that GR2PO significantly outperforms critic-free baselines that use immediate rewards and performs competitively against state-of-the-art actor-critic methods. Furthermore, GR2PO demonstrates competitive training efficiency. Inference tests on NVIDIA Jetson TX2 demonstrate the feasibility of deploying the learned policies on edge platforms. Further ablation experiments analyze the effects of parallel group size, return estimation methods, and target clipping ratio on learning performance. To support follow-up research, we will make the complete code publicly available after the paper is accepted, including the framework implementation, experimental configuration, and training and evaluation scripts.

Pengqin Wang, Qiming Zhang, Shaojie Shen +1

Robotics and Autonomous Systems Thrust, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China · Division of Emerging Interdisciplinary Areas, the Hong Kong University of Science and Technology, Hong Kong SAR, China · Intelligent Transportation Thrust, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China +1

Reinforcement Learning Group Relative Policy Optimization Critic-Free RL Robotic RL Continuous Control

Sep 14, 2026cs.LG

Refinement-based Flow Policy Optimization

Flow-based policies offer an expressive representation for online reinforcement learning, but conventional flow matching requires samples drawn from the distribution to be modeled. This poses a challenge when the desired action distribution is defined only implicitly by a Q-function, since directly sampling actions from the resulting distribution is generally intractable. We propose Refinement-Based Flow Policy Optimization (RFPO), a novel framework for training a flow policy in online reinforcement learning by alternating between Q-guided sample refinement and self-target flow matching. RFPO first generates actions from Gaussian noise using the current flow policy and then uses a finite-step stochastic refinement procedure to move them toward an energy-based distribution induced by the Q-function. Each refined action is then paired with its corresponding initial noise sample and used as a fixed target for flow-matching training. By repeatedly refining its own outputs and learning from the resulting targets, RFPO incorporates Q-guidance into the policy without requiring direct samples from the target distribution, while retaining the capacity to represent multiple action modes. We further provide a theoretical analysis of the distributional dynamics induced by RFPO. Across six continuous-control tasks, RFPO matches or outperforms a standard Gaussian-policy baseline on almost every task. Experiments on six synthetic two-dimensional target distributions with diverse geometries demonstrate that RFPO captures complex multimodal structure without mode collapse.

Bumgeun Park, Hyukjun Yang, Donghwan Lee

Korea Advanced Institute of Science and Technology (KAIST)

RL Control Flow Matching Policy Optimization Continuous Control

Sep 1, 2026cs.RO

On Global Regulatability of Robot Manipulators by Classical PID

A long-standing open problem in robot manipulator control is whether global regulation can be achieved by classical PID control. This paper provides an answer to this question for classical PID controllers with triple parameters (k_p,k_i,k_d) in R^3. We find and prove that for one-degree-of-freedom manipulators, the classical PID control guarantees global stability and asymptotic regulation under standard structural assumptions, and further derive explicit quantitative design conditions for the PID gains. However, for multi-degree-of-freedom cases, we can construct a robot manipulator satisfying the same structural assumptions for which no choice of PID gains (k_p,k_i,k_d) can achieve global asymptotic regulation. These results provide a fundamental understanding of the abovementioned open problem, revealing both the fundamental capability and intrinsic limitation of the classical PID control for robot manipulator dynamics.

Cheng Zhao, Jingru Zhu, Lei Guo

State Key Laboratory of Mathematical Sciences, AMSS, Chinese Academy of Sciences, Beijing 100190, China · School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China

Robotic Control Feedback Control Continuous Control

Aug 7, 2026cs.LG

Beyond Isolation: Unlocking Reinforcement Learning Component Synergy for Sample-Efficient Continuous Control

Reinforcement learning systems are significantly more complex than other machine learning paradigms due to inherent properties, causing RL system design to jointly account for many tightly coupled factors. Despite advances in individual algorithmic components, their functional interdependencies remain underexplored: do they exhibit mutual synergy or counterproductive interference? To bridge this gap, we conduct a systematic investigation and find that the efficacy of different components exhibits significant task-dependency, and naively stacking state-of-the-art techniques does not necessarily yield performance gains; instead, it often triggers emergent challenges, such as compounded non-stationarity. Building upon these findings, we distill a suite of actionable insights into the principled coordination of these components. Guided by these insights, we propose ROSER, an RL framework that coordinates three critical dimensions: Model-based Representation, Optimization Stability, and Experience Replay. Across diverse continuous-control benchmarks, ROSER consistently outperforms vanilla baselines and achieves 17.60% gains over naive stack. Our findings underscore the necessity of a holistic perspective in RL system design and paves the way for developing sample-efficient agents.

Qi Zhao, Guozheng Ma, Yilun Kong +9

Tsinghua University · Nanyang Technological University · Mila - Quebec Artificial Intelligence Institute +1

Reinforcement Learning Experience Replay Continuous Control Sample-Efficient RL

Aug 6, 2026cs.LG

Observation-Grounded Self-Predictive Reinforcement Learning for Visual Continuous Control

Sample-efficient policy learning from pixels is a long-standing challenge in reinforcement learning (RL). Recent dynamics-based representation learning methods have significantly improved the sample efficiency of model-free visual RL by learning dynamics-aware representations through auxiliary prediction performed either in latent space (self-prediction) or observation space (observation prediction). However, state-of-the-art methods from both categories still struggle on challenging visual control tasks when training data is limited. We posit that relying on either predictive objective alone may be insufficient. In contrast, observation prediction grounds learned representations in observation-level dynamics, but does not directly regularize the temporal predictability of latent representations over extended horizons. In this paper, we propose Observation-Grounded Self-Predictive Representations (OG-SPR), a model-free visual RL algorithm for continuous control that learns representations that are both temporally predictive in latent space and grounded in observation-level dynamics. OG-SPR incorporates two core auxiliary objectives: multi-step latent self-prediction and next-observation prediction. We empirically show that directly imposing latent self-prediction on the shared representation may over-constrain it and does not necessarily improve performance. To address this issue, OG-SPR introduces two lightweight adapters for latent self-prediction, allowing the shared representation to benefit from temporally predictive signals without being forced to directly satisfy the self-prediction objective. Experiments on 28 visual control tasks from the DeepMind Control Suite show that OG-SPR improves aggregate performance over state-of-the-art self-predictive and observation-predictive RL methods, with particularly pronounced gains in challenging domains such as dog and humanoid.

Xinwei Liu, Junyuan Liang, Jianting Zhang +1

Sun Yat-sen University · Purdue University

Predictive Representation Learning Continuous Control Model-Free RL Visual RL

Aug 5, 2026cs.LG

Learning When to Stop: Prefix-Optimal Dynamic Diffusion Policies for Continuous Control

Diffusion policies are a powerful policy class for continuous control, but their iterative denoising process creates a substantial computational bottleneck. Reducing this cost requires adapting the number of denoising steps to the difficulty of each action while preserving task performance. We introduce Prefix-Optimal Generative Policies (POGP), a framework that learns a prefix value function at every intermediate denoising step through a Bellman-style recursion over the denoising chain. The prefix value function serves two purposes: it provides an auxiliary training objective that encourages intermediate outputs to become high-quality actions, and it enables a test-time stopping rule that terminates denoising when additional steps are unlikely to produce meaningful improvement. Across four MuJoCo environments and comparisons with 12 baselines, POGP reduces the required number of denoising iterations by approximately 2.7-fold while retaining near-full task performance. Compared with state-of-the-art dynamic diffusion baselines, prefix training also improves final task performance by approximately 3.5%. These results indicate that supervising intermediate denoising steps is useful not only for adaptive early stopping, but also as an auxiliary objective that improves the learned policy.

Rohit Kumar Salla, Manoj Saravanan, Simon Stepputtis

Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia, USA

Diffusion Policy Continuous Control RL for Diffusion Models Policy Learning Diffusion Models

Jul 30, 2026eess.SY

Safety-Gated Agentic Supervisory Control on a Coupled Distillation Benchmark: Regime Map, Auditable Gate, and Co-Design Findings

An open-weight LLM can write composition setpoints every five minutes. What a plant still needs is a hard check: named constraints, logged margins, and an admit/block decision before the regulatory layer moves. This paper puts that check in a rule-based forked-twin counterfactual gate (nine pinned constraints) and leaves the regulatory layer unchanged. On Skogestad's Column A the ladder is PID-only (C0), linear MPC (C1), ungated agent (C2), and gated agent (C3) under one contract: identical level closure (M_D, M_B), scenarios, and seeds; C2/C3 share the linear-MPC backend. The split is not subtle. Off-nominal target acquisition: the agent beats Pareto-tuned linear MPC in the strong band (C2/C1 IAE ratio 0.361 at the upper CI). Disturbance rejection on the same 16-point grid inverts by 16.03 at the upper CI (10.18 at the point estimate), where an ungated LLM supervisor does not belong. The gate compresses a specification-abandonment attractor into a bounded offset (d approx. -1.4; P95 cell IAE 11.5 to 0.77). A one-line prompt fix removes the attractor at source (6/10 to 0/10; sensitivity only, not a new headline). In a 250-cell statistical pass, 534 of 590 gate interventions are spec-on-bound geometry: the operating specification sits on a safety limit, so a well-behaved OP becomes inoperable while misbehaving ones are only contained; 318 blocks still correct actively harmful proposals. Headlines are single-column and model-conditional on DeepSeek-V4-Flash. A second-family sweep (NVIDIA Nemotron-3-Super) keeps the disturbance-rejection fails band and plant-side failure geography; magnitudes and protocol operability stay model-conditional, and Super target-acquisition strong cells are survivors only (not confirmation). Transfer means twin, constraint envelope, and setpoint interface, not a second plant class measured here.

Christian Rosenthal

Sole author — independent research, industrial AI methodology

Language Model-Based Control Safety-Critical Control Feedback Control Continuous Control Runtime Assurance

Jul 29, 2026cs.LG

Do You Really Need to Pretrain Q-Functions for Online RL Fine-Tuning?

Pre-training followed by fine-tuning has become the dominant recipe for learning performant policies, and in value-based reinforcement learning (RL) this raises a natural question: given a pretrained policy, should the Q-function be pretrained on offline data too? Conventional wisdom suggests it should, but recent results show that online RL with a randomly-initialized Q-function can result in highly performant and reliable policies without needing to pretrain the Q-function. In this paper, we systematically study whether pretraining the Q-function actually helps when fine-tuning on top of a pretrained base policy. We find, surprisingly, that naive Q-function pretraining often provides little benefit over random initialization. We show this stems from a fundamental mismatch: the Q-function learned during pretraining targets the pretrained policy's Q-function, not the Q-function that online fine-tuning converges to, and this gap persists even after offline value maximization. Motivated by this finding, we propose Initialization via Policy Ensemble (IPE), a simple method that trains multiple diverse policies and uses their pooled rollouts to bootstrap the Q-function learning in online RL. Across a suite of challenging continuous control benchmarks, IPE yields an average 1.26x improvement in fine-tuning performance over naive Q-function pre-training.

Perry Dong, Ron Polonsky, Dorsa Sadigh +1

Stanford University

Continuous Control RL Fine-Tuning Policy Learning Offline-To-Online RL

Jul 29, 2026cs.HC

Sensor-Placement-Agnostic Sonomyography: Toward Continuous High-Dimensional Control by Users with Tetraplegia

Sonomyography (SMG) enables continuous device control via ultrasound-measured muscle deformation signals, but existing SMG interfaces generally require substantial user- and sensor-location-specific training data and provide only one proportional signal or task-specific classification. We present a real-time, sensor-placement-agnostic SMG control system based on sparse optical flow tracking that enables continuous 1-DOF control after minimal calibration (3 pose definitions). We also present a preliminary expansion of this method that augments this algorithm with a short computer-aided calibration to enable 2-DOF control. We evaluate both 1- and 2-DOF systems' performance for a preliminary cohort of 3 cervical spinal cord injury survivors and 6 uninjured individuals across 6 sensor placements spanning the arm, neck, and upper torso. As assessed by a cursor trajectory tracking task, all participants achieved continuous 1-DOF control at all tested sensor locations (even those that relied on passive tissue motions), with all participants achieving <5.5% tracking error using at least one placement (and many <4% across many). All participants were also able to modulate 2D cursor position via the 2-DOF system, with varying levels of control authority, and several were able to complete a 2D drawing task, constituting the first (to our knowledge) demonstration of location-agnostic multi-DOF continuous SMG-based control. These results highlight the promise of SMG to enable rapidly calibratable, high-dimensional, sensor-placement-agnostic device control by users with tetraplegia, and also illuminate key challenges in both signal processing and practical system deployment. To enable further development by scientific and user communities, developed algorithms have been open-sourced as part of the OpenMyoControl project on SimTK (simtk.org/projects/openmyocontrol).

Gavin Sueltz, Vikram Athithan, Emma Ferran +3

Departments of Mechanical Engineering and Physical Medicine & Rehabilitation, the Kahlert School of Computing, and the Robotics Center at the University of Utah, Salt Lake City, UT 84112, USA

Wearable Sensing Assistive Robotics Continuous Control

Jul 6, 2026cs.AI

Graph Sparse Sampling: Breaking the Curse of the Horizon in Continuous MDP Planning

Planning under uncertainty in continuous domains is essential for autonomous systems, yet computationally demanding. Tree-based search methods such as Monte Carlo Tree Search (MCTS) remain popular, but their branching structure can require sampling budgets that grow exponentially with lookahead depth in the worst case. From a tree perspective, continuous state or action spaces become especially challenging, since the planner must decide where to search in an infinite branching hierarchy. We propose Graph Sparse Sampling (GSS), an online planning algorithm that shares sampled futures across many candidate decisions, rather than sampling separate successors for each candidate action. This branch-free graph exposes large GPU-friendly batches, while using heuristics to focus computation. We prove finite-sample performance guarantees for GSS covering full-rank or low-rank generative simulators via smoothed backups, and discrete or sampled continuous action spaces. Under suitable overlap, regularity, and action-coverage conditions, these bounds have polynomial dependence on the planning horizon, formalizing when shared futures can avoid the exponential horizon dependence of tree-shaped sparse sampling. We demonstrate continuous-control simulations where GSS substantially outperforms tree-based planners on long horizons or achieves near-optimal performance, supporting no-branching graph planning as a complementary design principle for online control.

Idan Lev-Yehudi, Vadim Indelman

Technion Autonomous Systems Program (TASP), Technion – Israel Institute of Technology · Stephen B. Klein Faculty of Aerospace Engineering, Technion – Israel Institute of Technology · Faculty of Data and Decision Sciences, Technion – Israel Institute of Technology

Adaptive Sampling Graph-Based Planning Model-Based Planning Continuous Control

Jun 21, 2026cs.LG

Low-power analogue neural networks with trainable nonlinear connections for continuous control

Physical neural networks promise low-power machine learning by computing directly with analogue device physics, but most architectures force nonlinear device responses to act as scalar weights. Inspired by Kolmogorov-Arnold networks, we place trainable nonlinear functions on the connections, making each physical connection a learnable computational element. Realising these functions as analogue band-pass filters on field-programmable analogue arrays, we find that the benefit is task-dependent and follows from the smoothness of the physical basis: the networks represent smooth, continuously valued targets, including robotic kinematics, continuous control, and photovoltaic maximum-power-point tracking, with far fewer nodes and connections than multilayer perceptrons, but offer no parameter-efficiency advantage on classification-like decision boundaries. Trained networks transfer to hardware across approximately 35,000 connections with quantified fidelity, and a dedicated CMOS implementation is projected to operate at approximately 30 microwatts. A memristive realisation reproduces the same behaviour in simulation, indicating that the advantage comes from placing trainable nonlinearity on connections, rather than from a particular device.

Ian T. Vidamour, Fernando Aguirre, Thomas J. Hayward +13

School of Computer Science, University of Sheffield, Sheffield, S1 4DP, United Kingdom · Intrinsic Semiconductor Technologies, London, United Kingdom · School of Chemical, Biological, and Materials Science Engineering, University of Sheffield, Sheffield, S1 3JD, United Kingdom +5

Efficient Neural Network Inference Kolmogorov-Arnold Networks Continuous Control Neuromorphic Computing

Jun 19, 2026cs.LG

Backpropagating Through Simulation: Analytic Policy Gradients for Sample and Learning Efficient Differentiable Continuous Control

Model-free reinforcement learning algorithms such as Proximal Policy Optimization (PPO) treat the environment as a black box, estimating policy gradients from sampled rewards; this process demands millions of interactions and relies on high-variance advantage estimates. When environment dynamics are differentiable, the return is an end-to-end differentiable function of the policy parameters, enabling exact gradient computation via backpropagation through simulation. We term this approach Analytic Policy Gradients (APG) and evaluate it against PPO on four continuous control tasks of increasing dynamical complexity: a one-dimensional point-mass target-reaching task, a 2D point-mass navigation task with obstacle avoidance, a 2D rigid-body T-block pushing task, and a 7-DOF Franka FR3 end-effector reaching task. Both algorithms share identical model architectures, observation normalization, and optimizer settings. To decouple sample efficiency from compute efficiency, we design a multi-axis evaluation protocol that records performance against environment steps and gradient steps. We report a segmented backpropagation scheme with MC and critic-based bootstrap modes that mitigates gradient degradation on long-horizon tasks, and present ablations over segment length and bootstrap strategy.

Yueci Deng

School of Data Science, The Chinese University of Hong Kong, Shenzhen

Robotic RL Policy Gradient Methods Continuous Control Policy Learning Sample-Efficient RL

Jun 11, 2026eess.SY

Interaction Dynamics MPC for Knee Rehabilitation Exoskeletons: A Closed-Loop SEA Outer-Loop Study

Safe rehabilitation is an interaction-dynamics problem: the controller must regulate a prescribed motion while absorbing involuntary spasm, voluntary effort, actuator compliance, and model mismatch as disturbances. This paper instantiates the predictive interaction-dynamics framework of the base pHRI formulation on a SEA knee joint. SEA feedforward reduces the gravity-compensated knee to the same scalar double integrator as the base framework, while a dynamic-residual measurement from spring deflection supplies an interaction-disturbance observation. A steady-state target converts the estimated disturbance into a cancelling input, and a finite-horizon quadratic program regulates deviations from that target under range-of-motion, torque, and velocity constraints. The evaluation matches stiffness and damping across controllers so gains cannot be attributed to higher impedance. Under a motion-opposing

15\unit{Nm}

step, classical impedance and MPC without estimation produce about

500\unit{mrad}

steady-state error, whereas Kalman-augmented interaction MPC reduces this to

1.17\unit{mrad}

at 100~~Hz and $0.70\unit{mrad}$ at 500~~Hz; the 500Hz peak is $7.27\unit{mrad}$ . In 30 randomized trials, the 95th-percentile peak is $21.57\unit{mrad}$ . Bounded Assist-as-Needed scheduling, a corrective-channel energy tank, constrained OSQP stress cases, direct MuJoCo execution, and a posture-clamped MyoSuite knee slice are implemented. The framework holds on a single-mass, closed-inner-loop SEA approximation; an explicit two-mass plant with a finite-bandwidth, pole-placed inner torque loop (SectionVIII) confirms this for nominal tracking but shows delivered torque can overshoot the commanded bound by 21.7% near saturation. Scope excludes clinical intent recognition, full-system passivity, safety certification, hardware trials, and multi-joint validation.

Yongyan Cao, Jinshan Tang

Voryx Robotics · George Mason University, Dept. of Health Administration and Policy

Robotic Control Model Predictive Control Assistive Robotics Continuous Control Human-Robot Interaction

Jun 7, 2026cs.AI

Self-Evolving Scientific Agent Designs Physically Reasoned White-Box Fluid Control

While neural networks excel in autonomous control, their black-box nature makes control decisions difficult to interpret and diagnose in dynamic fluids. Here, we show how self-evolving scientific agents can design explicit, neural-network-free white-box controllers by iteratively interpreting simulation evidence, accumulating control knowledge and refining controller code. We demonstrate this approach on an underactuated two-joint swimmer navigating unsteady flows via joint angular accelerations. Starting from a target-blind propulsive controller, the agent gradually constructs key mechanisms, including travelling-wave propulsion, body-frame guidance, phase-selective steering, redirect bursts and adaptive relief. The resulting controllers reach targets and generalize across changes in target position, wake geometry, cylinder count, and inflow speed without revision. Moreover, 2D control priors transfer successfully to accelerate 3D adaptation. Our work demonstrates that self-evolving agents can autonomously design physically reasoned and generalizable white-box fluid control, showing a promising paradigm beyond traditional reinforcement learning and black-box neural network control.

Boai Sun, Wenjin Guo, Zongmin Yu +1

National University of Singapore, Singapore.

Robotic Control Continuous Control Autonomous Scientific Discovery Self-Evolving Agents

Jun 6, 2026cs.RO

PRISM: PRior-guided Imagination Sampling in world Models

A learned world model provides a powerful physical intuition for evaluating future states. But its effectiveness in continuous control also depends critically on how candidate actions are generated for model-based planning. Rather than solely asking how accurately a model can simulate the future, we ask: which candidate actions are worth evaluating in the first place? Existing planners typically search arbitrarily or use expert demonstrations only to initialize a sampling mean, discarding the expert's state-conditioned confidence. Properly guiding this search requires a robust action prior, yet current approaches often rely on independent visual encoders or large-scale VLMs to obtain one. We argue that this architectural bloat is unnecessary: the exact same data - and the learned representations of the world model itself - inherently encode the agent's action intuition. We introduce PRISM, a task-agnostic framework that extracts both from a single dataset while maintaining strict architectural simplicity. Building on a standard JEPA-style latent world model, PRISM attaches a lightweight MLP directly to its frozen encoder to predict a state-conditioned Gaussian prior. At plan time, PRISM fuses this prior into the planner's sampling distribution via a precision-weighted Product-of-Gaussians update. This parameter-free, closed-form integration steers the sampling process, making the prior confident where it is and ceding control where it is not. PRISM improves success rates by 35 percentage points over vanilla world-model-based MPC on Cube and 32 percentage points on PushT, without introducing significant inference overhead.

Yuhai Wang, Jiawei Xia, Rongxuan Zhou +4

1Northeastern University · University of California, Berkeley · 3Qiyuan Lab +1

World Models for Robotics World Model-Based Planning Model-Based Planning Continuous Control Latent World Models

Jun 4, 2026cs.LG

Representation Learning Enables Scalable Multitask Deep Reinforcement Learning

Scaling reinforcement learning (RL) to diverse multitask settings remains a central challenge. While recent advances in model-based RL achieve strong performance, they rely on planning and complex training pipelines, making it unclear which components are essential for scalability. We revisit this question and argue that the primary driver of scalable multitask RL is not model-based control, but \emph{representation learning}. In particular, we show that combining predictive, model-based representations with high-capacity value function approximation is sufficient to achieve strong performance, even without planning. We evaluate a simple model-free algorithm, MR.Q, coupled with auxiliary predictive objectives into a scalable actor-critic architecture. This approach outperforms a recent world-model-based method and a range of deep RL baselines across a diverse suite of multitask continuous control tasks, while significantly reducing computational overhead and improving wall-clock efficiency. We observe consistent improvements with increased model capacity and show through ablations that predictive representation learning is critical for performance.

Johan Obando-Ceron, Lu Li, Scott Fujimoto +3

Mila – Québec AI Institute · Université de Montréal · McGill University +2

Reinforcement Learning Representation Learning for RL Continuous Control Multi-Task RL Model-Free RL

May 22, 2026cs.LG

Reflex: Reinforcement Learning with Reflection Symmetry Exploitation in State-Based Continuous Control

Reinforcement learning has long struggled with poor sample efficiency. One promising approach to mitigate this problem is leveraging group-invariant Markov Decision Processes (

G

-invariant MDPs). Existing works in this direction have primarily focused on image-based RL and rotational symmetry such as

\mathrm{SO(2)}

, leaving state-based RL and reflection symmetry largely underexplored. In this work, we focus on state-based continuous control tasks and exploit reflection symmetry by introducing Reflex, a paradigm that seamlessly integrates with both on-policy and off-policy RL algorithms. We formalize two types of reflection-axial reflection and bilateral reflection, and characterize their corresponding transformations. Building on a theoretical analysis of symmetry-preserving optimal value functions and policies, Reflex integrates reflection symmetry into policy learning through principled symmetry regularization mechanisms. We integrate Reflex with PPO and SAC, and evaluate it on a suite of OpenAI Gym and DeepMind Control benchmarks, demonstrating superior performance over standard baselines while improving sample efficiency. Our code is available at https://github.com/TonyStark042/Reflex.

Shuai Zhen, Yifan Zhang, Yuling Wang +1

Beijing University of Posts and Telecommunications · Hangzhou Dianzi University

RL Control Reinforcement Learning Continuous Control Sample-Efficient RL

May 20, 2026cs.LG

PREFINE: Preference-Based Implicit Reward and Cost Fine-Tuning for Safety Alignment

We address the problem of making a pre-trained reinforcement learning (RL) policy safety-aware by incorporating cost constraints without retraining it from scratch. While costs could be numerically encoded, we assume a more general setting is when costs are provided as preferences. Given a reward-optimized policy and a small dataset of preferred (low-cost) and dispreferred (high-cost) trajectories, our goal is to fine-tune the policy to generate low-cost behaviors while retaining high rewards. Unlike standard RLHF in language models, where preferences are defined over responses to the same prompt, our setting involves trajectory-level preferences in continuous control environments. We introduce PREFINE: Preference-based Implicit Reward and Cost Fine-Tuning for Safety Alignment which is a preference-based fine-tuning method that adapts Direct Preference Optimization (DPO), which is now widely used for LLM fine-tuning, to the sequential decision making setting. PREFINE constructs policy-sampled counterfactual trajectories to establish meaningful preference contrasts and jointly optimizes for reward retention and safety alignment. Empirically, PREFINE reduces constraint violations and catastrophic failures by over 60% while maintaining original reward behavior. PREFINE produces policies that achieve low-cost, high-reward performance with significantly improved data and computational efficiency compared to full offline RL or imitation learning, bridging preference alignment and safe policy adaptation in continuous domains.

Richa Verma, Bavish Kulur, Sanjay Chawla +1

TCS Research, Department of CSE, IIT Madras India · Department of Computing Science, University of Alberta Canada · Qatar Computing Research Institute, Hamad Bin Khalifa University Qatar +1

Direct Preference Optimization Constrained RL Continuous Control Safe RL RL Fine-Tuning Preference-Based RL

May 19, 2026cs.RO

Implicit Action Chunking for Smooth Continuous Control

Reinforcement learning often produces high-frequency oscillatory control signals that undermine the safety and stability required for physical deployment. Explicit action chunking addresses this by predicting fixed-horizon trajectories but scales the policy output dimension proportionally with the horizon length, leading to optimization difficulties and incompatibility with standard step-wise interaction. To overcome these challenges, this paper proposes Dual-Window Smoothing (DWS), an implicit action chunking framework for smooth continuous control. Unlike explicit methods, DWS enforces temporal coherence without expanding the action space. It uses a dual-window design: an execution window that ensures physical smoothness through deterministic modulation, and a value window that aligns temporal-difference targets over the horizon to correct critic bias caused by open-loop execution. DWS also includes a lightweight actor-side temporal regularizer based on first-order action differences to promote global continuity. This design effectively bridges the gap between temporal abstraction and reactive step-wise control. Experiments on benchmarks including the DeepMind Control Suite and industrial energy management tasks show that DWS outperforms state-of-the-art (SOTA) baselines. In complex vision-based autonomous driving tasks, DWS achieves smoother control, safer behavior with reduced jitter, and attains a 100% success rate.

Bosun Liang, Shuo Pei, Zirui Chen +5

Department of Data and Systems Engineering, The University of Hong Kong, Hong Kong SAR, China · Beijing Institute of Technology, Zhuhai, China · College of Computer Science, Sichuan University, Chengdu, China

RL Control Continuous Control Action Chunking

May 15, 2026cs.LG

EfficientTDMPC: Improved MPC Objectives for Sample-Efficient Continuous Control

We introduce EfficientTDMPC, a sample-efficient model-based reinforcement learning method for continuous control built on the TD-MPC family of algorithms. Central to this family is a planner that aims to find an action sequence that maximizes the estimated return. The return is estimated using a learned model and value networks, each of which can introduce error. EfficientTDMPC introduces three contributions that improve performance by aiming to reduce this error. First, we introduce a multi-horizon planning objective that evaluates the value at different rollout depths and averages them. Second, to our knowledge we are the first to train a value-equivalent dynamics ensemble. Our improved objective then averages over rollouts from multiple dynamics heads. Third, we add pessimistic reanalyze for tasks that can terminate early. Applying our contributions to a recent baseline (BMPC) yields EfficientTDMPC, which to our knowledge is the new state of the art in sample efficiency on HumanoidBench and the DeepMind Control Suite, reaching BMPC's final aggregated performance using 57% fewer environment steps.

Thomas Evers, Cristian Meo, Wendelin Bohmer +2

TU Delft · LatentWorlds AI

Model Predictive Control Continuous Control Sample-Efficient RL Model-Based RL

May 14, 2026physics.plasm-ph

Real-time virtual circuits for plasma shape control via neural network emulators

Reliable position and shape control in tokamak plasmas requires accurate real-time regulation of several strongly coupled shape parameters. The control vectors that disentangle these couplings, referred to as \textit{virtual circuits} (VCs), enable independent shape parameter control for a specific Grad--Shafranov (GS) equilibrium. Numerical calculation of VCs is not currently feasible in real time, therefore VCs are usually computed prior to each experiment, using a small number of reference GS equilibria sampled along the desired scenario trajectory, with each VC used to control the plasma within a preset time interval. While effective near the reference equilibrium, this approach can lead to degraded performance as the plasma departs from the reference equilibrium and/or from the desired trajectory, and it complicates the design of robust control strategies for rapidly evolving plasma configurations. In this paper, we construct neural-network-based emulators of plasma shape parameters from which VCs can be derived, to provide the MAST Upgrade (MAST-U) plasma control system with state-aware VCs in real-time. To do this, we develop an extensive library of over a million simulated GS equilibria, covering a substantial portion of the MAST-U operational space. These emulators provide differentiable functions whose gradients can be rapidly computed, enabling the derivation of accurate VCs for real-time shape control. We perform extensive verification of the emulated VCs by testing whether they disentangle the control problem. The neural-network-based approach delivers high accuracy and orthogonality across a diverse range of equilibria. This work establishes the physical validity of emulated VCs as a scalable and general alternative to schedules of precomputed VCs.

Alasdair Ross, George K. Holt, Kamran Pentland +7

STFC Hartree Centre, Sci-Tech Daresbury, Keckwick Lane, Daresbury, Warrington, WA4 4AD, United Kingdom · United Kingdom Atomic Energy Authority, Culham Campus, Abingdon, Oxfordshire, OX14 3DB, United Kingdom

Neural Surrogate Modeling Feedback Control Continuous Control

May 12, 2026cs.LG

Debiased Model-based Representations for Sample-efficient Continuous Control

Model-based representations recently stand out as a promising framework that embeds latent dynamics information into the representations for downstream off-policy actor-critic learning. It implicitly combines the advantages of both model-free and model-based approaches while avoiding the training costs associated with model-based methods. Nevertheless, existing model-based representation methods can fail to capture sufficient information about relevant variables and can overfit to early experiences in the replay buffer. These incur biases in representation and actor-critic learning, leading to inferior performance. To address this, we propose Debiased model-based Representations for Q-learning, tagged DR.Q algorithm. DR.Q explicitly maximizes the mutual information between the representations of the current state-action pair and the next state besides minimizing their deviations, and samples transitions with faded prioritized experience replay. We evaluate DR.Q on numerous continuous control benchmarks with a single set of hyperparameters, and the results demonstrate that DR.Q can match or surpass recent strong baselines, sometimes outperforming them by a large margin. Our code is available at https://github.com/dmksjfl/DR.Q.

Jiafei Lyu, Zichuan Lin, Scott Fujimoto +5

1Tencent Hunyuan · 2McGill University · School of Computer

Representation Learning Continuous Control Actor-Critic Methods Sample-Efficient RL Off-Policy RL Model-Based RL

May 9, 2026cs.RO

Beyond Self-Play: Hierarchical Reasoning for Continuous Motion in Closed-Loop Traffic Simulation

Closed-loop traffic simulation requires agents that are both scalable and behaviorally realistic. Recent self-play reinforcement learning approaches demonstrate strong scalability, but their equilibrium strategies fail to capture the socially aware behaviors of real human drivers. We propose a hierarchical architecture that goes beyond self-play by combining high-level multi-agent interaction reasoning with low-level continuous trajectory realization. Specifically, a Stackelberg-style Multi-Agent Reinforcement Learning (MARL) module generates interaction-aware intention commands. These commands condition a low-level continuous motion module, translating the strategic intent into physically consistent, scene-responsive control sequences. To mitigate distribution shift in closed-loop deployment, we introduce a hybrid co-training scheme combining MARL with auxiliary recovery supervision. Experiments on a SUMO-based urban network demonstrate that the proposed framework achieves superior control smoothness and safety compared to self-play and passive imitation baselines, while maintaining competitive traffic efficiency.

Weifan Zhang, Xiaofeng Zhao, Adel Bazzi +3

Isuzu Technical Center of America, Plymouth, MI, USA · School of Aeronautics and Astronautics, Purdue University, West Lafayette, IN 47907, USA

Human Behavior Simulation Hierarchical RL Continuous Control Autonomous Driving Simulation Multi-Agent Reinforcement Learning

May 9, 2026cs.AI

When (and How) to Trust the Expert: Diagnosing Query-Time Expert-Guided Reinforcement Learning

Many continuous-control problems ship with a competent but suboptimal controller (a tuned PID, a hand-designed gait). A growing family of methods uses such controllers as queryable experts during RL, but each method has been proposed in isolation, on a different benchmark, without imperfect-expert testing. We harmonize the comparison on a shared SAC backbone, common HPO and evaluation protocols, 100/50 seeds per (env, method), and a degradation sweep over expert undertuning, action bias, and observation noise. The comparison surfaces three failure modes single-paper evaluations miss: (F1) a critic blind spot under argmax-plus-bootstrap that drags IBRL below no-expert SAC on experts close to the no-expert-RL ceiling (RL-near-ceiling, distinct from the absolute physical ceiling); (F2) residual saturation on far-from-optimal experts; and (F3) warm-start buffer poisoning that collapses training-time-handoff methods under deployment-time expert undertuning. No single method dominates: each wins on one task-structure regime and fails predictably elsewhere; on RL-near-ceiling experts (FourTank, GlassFurnace) no query-time method clears the expert within our 1M-step budget, leaving open whether this is a fundamental wall or a budget effect. We convert the spread into a testable decision rule keyed on three pre-training observables (expert quality, task termination, perturbation type). The benchmark, taxonomy, and decision rule are the primary contribution; we additionally describe EDGE, a softmax-over-ensemble-LCB design point used to demonstrate that both axes the taxonomy points to (gate form, scoring rule) are individually exploitable.

Yann Berthelot, Philippe Preux, Riad Akrour

UMR 9189 – CRIStAL · CNRS, Inria, Centrale Lille · Saint-Gobain Research Paris +3

RL Control RL Benchmarks Continuous Control Robust RL

May 9, 2026cs.LG

Generative Actor-Critic with Soft Bridge Policies

Expressive generative policies such as diffusion and flow models are appealing for MaxEnt online reinforcement learning because of their ability to model multimodal and highly non-Gaussian action distributions. However, training effective soft generative policies faces two obstacles that often arise together. First, marginal action densities are often unavailable, so existing methods typically rely on entropy bounds, heuristic proxies or approximations. Second, iterative shared-parameter samplers raise inference cost and require backpropagation through time over repeated network evaluations, increasing memory cost and destabilizing policy optimization. These obstacles motivate us to seek a generative policy that exposes a tractable MaxEnt objective while requiring only a single sampled actor forward pass for action generation. To this end, we propose soft generative actor-critic (SoftGAC), whose actor defines a stochastic bridge from a fixed base latent to a terminal action latent in pre-tanh space. This structured bridge allows us to lift the MaxEnt objective as an analytically tractable path-wise relative-entropy objective against a high-entropy reference process. In practical finite-step implementation, this relative entropy reduces exactly to sampled transition control energy and thus provides principled soft regularization. Moreover, we keep the single-pass actor lightweight by using small step-specific bridge transitions, each evaluated only once per sampled action, while maintaining a parameter budget comparable to strong actor baselines. Extensive experiments on challenging continuous-control benchmarks show that SoftGAC attains higher or competitive returns than strong generative policy baselines, including diffusion and flow-matching policies, while staying in the low-latency regime of one-pass actors and showing considerable improvements in the compute-return tradeoff.

Ke He, Le He, Shunpu Tang +2

Guangzhou University · Zhejiang University · Southeast University

Policy Optimization Maximum Entropy RL Continuous Control RL for Generative Models

May 2, 2026cs.RO

Bi-Level Reinforcement Learning Control for an Underactuated Blimp via Center-of-Mass Reconfiguration

This paper investigates goal-directed tracking control of underactuated blimps with center-of-mass (CoM) reconfiguration. Unlike conventional overactuated blimp designs that rely on redundant actuation for simplified control, this paper focuses on a compact architecture consisting of two thrusters and a movable internal slider, aiming to improve energy efficiency and payload capacity. This hardware-efficient configuration introduces significant underactuation and strong nonlinear coupling between CoM dynamics and vehicle motion. To address these challenges, this paper proposes a bi-level reinforcement learning framework that explicitly decouples task-level CoM planning from continuous thrust control. The outer policy determines a target-dependent CoM configuration prior to flight, while the inner policy generates thrust commands to track straight-line references. To ensure stable learning, this paper introduces a two-stage learning strategy, supported by a convergence analysis of the resulting bi-level process. Extensive simulations and real-world experiments on a 27-goal evaluation set demonstrate that the proposed method consistently outperforms fixed-CoM baselines and PID-based controllers, achieving higher tracking accuracy, enhanced robustness, and reliable sim-to-real transfer.

Xiaorui Wang, Hongwu Wang, Yue Fan +2

RL Control Aerial Robotics Hierarchical RL Continuous Control

May 1, 2026cs.LG

SAVGO: Learning State-Action Value Geometry with Cosine Similarity for Continuous Control

While representation and similarity learning have improved the sample efficiency of Reinforcement Learning (RL), they are rarely used to shape policy updates directly in the action space. To bridge this gap, a geometry-aware RL algorithm that explicitly incorporates value-based similarity into the policy update, State-Action Value Geometry Optimization (SAVGO), is proposed. In detail, SAVGO learns a joint state-action embedding space in which pairs with similar action-value estimates exhibit high cosine similarity, while dissimilar pairs are mapped to distinct directions. This learned geometry enables the generation of a similarity kernel over candidate actions sampled at each update, allowing policy improvement to be guided directly toward higher-value regions beyond local gradient-based updates. As a result, representation learning, value estimation, and policy optimization are unified within a single geometry-consistent objective, while preserving the scalability of off-policy actor-critic training. The proposed method is evaluated on standard MuJoCo continuous-control benchmarks, demonstrating improvements over strong baselines on challenging high-dimensional tasks. Ablation studies are done to analyze the contributions of value-geometry learning and similarity-based policy updates.

Stavros Orfanoudakis, Pedro P. Vergara

Delft University of Technology, The Netherlands

Representation Learning Policy Optimization Continuous Control Actor-Critic Methods

Apr 28, 2026cs.AI

Multi-action Tangled Program Graphs for Multi-task Reinforcement Learning with Continuous Control

Over the past few decades, machine learning has been widely used to learn complex tasks. Reinforcement Learning (RL), inspired by human behavior, is a great example, as it involves developing specific behaviours for specific tasks. To further challenge algorithms, Multi-Task RL (MTRL) environments have been introduced, requiring a single model to learn multiple behaviors. The Tangled Program Graph (TPG) algorithm is a Genetic Programming (GP) algorithm designed for discrete MTRL environments. Recently, the MAPLE algorithm has been proposed, as another GP algorithm that achieves high results in single task continuous RL environments. A variation of the TPG is proposed alongside MAPLE, named Multi-Action TPG (MATPG) that aggregates MAPLE agents, and creates a control flow to activate them. Initially tested on single task RL environments only, MATPG achieved similar results to MAPLE. In this work, we present a new benchmark based on the MuJoCo Half Cheetah from Gymnasium. This benchmark features five distinct obstacles that are randomly positioned in front of the agent, each of which demands a unique behavior. This benchmark serves as a use case for MATPG, to prove its ability as a GP solution for continuous MTRL environments. Our experiments demonstrate its superiority in this multi-task use case when combined with lexicase selection. Furthermore, we examine the interpretability of the evolved graph, revealing that the decision flow of the model is fully interpretable.

Quentin Vacher, Nicolas Beuve, Mickaël Dardaillon +1

IETR · Univ Rennes, INSA Rennes, CNRS, IETR – UMR 6164, F-35000 Rennes, France

RL Benchmarks Genetic Programming Continuous Control Multi-Task RL

Apr 16, 2026cs.LG

Mean Flow Policy Optimization

Diffusion models have recently emerged as expressive policy representations for online reinforcement learning (RL). However, their iterative generative processes introduce substantial training and inference overhead. To overcome this limitation, we propose to represent policies using MeanFlow models, a class of few-step flow-based generative models, to improve training and inference efficiency over diffusion-based RL approaches. To promote exploration, we optimize MeanFlow policies under the maximum entropy RL framework via soft policy iteration, and address two key challenges specific to MeanFlow policies: action likelihood evaluation and soft policy improvement. Experiments on MuJoCo, DeepMind Control Suite and HumanoidBench benchmarks demonstrate that our method, Mean Flow Policy Optimization (MFPO), achieves performance comparable to or exceeding current diffusion-based baselines while considerably reducing training and inference time. Our code is available at https://github.com/dongxiaoyi-xyz/MFPO.

Xiaoyi Dong, Xi Sheryl Zhang, Jian Cheng

C2DL, Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · AiRiA +1

MeanFlow Mean Flow Continuous Control Policy Learning