# FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control

arXiv: https://arxiv.org/abs/2603.12612 · HTML: https://mlsanity.com/paper/2603.12612/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 35 references that are in the map.
- First posted on arXiv 13 Mar 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 25 of 25)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [A Distributional Perspective on Reinforcement Learning](https://mlsanity.com/paper/1707.06887/) (2017)
- [Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning](https://mlsanity.com/paper/2109.11978/) (2021)
- [TD-MPC2: Scalable, Robust World Models for Continuous Control](https://mlsanity.com/paper/2310.16828/) (2023)
- [Efficient Online Reinforcement Learning with Offline Data](https://mlsanity.com/paper/2302.02948/) (2023)
- [Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments](https://mlsanity.com/paper/2301.04195/) (2023)
- [Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics](https://mlsanity.com/paper/2005.04269/) (2020)
- [HumanoidBench: Simulated Humanoid Benchmark for Whole-Body Locomotion and Manipulation](https://mlsanity.com/paper/2403.10506/) (2024)
- [Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control](https://mlsanity.com/paper/2405.16158/) (2024)
- [CrossQ: Batch Normalization in Deep Reinforcement Learning for Greater Sample Efficiency and Simplicity](https://mlsanity.com/paper/1902.05605/) (2019)
- [Distributional Soft Actor-Critic: Off-Policy Reinforcement Learning for Addressing Value Estimation Errors](https://mlsanity.com/paper/2001.02811/) (2020)
- [Simplifying Deep Temporal Difference Learning](https://mlsanity.com/paper/2407.04811/) (2024)
- [MuJoCo Playground](https://mlsanity.com/paper/2502.08844/) (2025)
- [FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control](https://mlsanity.com/paper/2505.22642/) (2025)
- [Distributional Soft Actor-Critic with Three Refinements](https://mlsanity.com/paper/2310.05858/) (2023)
- [Learning Sim-to-Real Humanoid Locomotion in 15 Minutes](https://mlsanity.com/paper/2512.01996/) (2025)
- [Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents](https://mlsanity.com/paper/2510.13704/) (2025)
- [Coarse-to-fine Q-Network with Action Sequence for Data-Efficient Reinforcement Learning](https://mlsanity.com/paper/2411.12155/) (2024)
- [TDMPBC: Self-Imitative Reinforcement Learning for Humanoid Robot Control](https://mlsanity.com/paper/2502.17322/) (2025)
- [Natural and Robust Walking using Reinforcement Learning without Demonstrations in High-Dimensional Musculoskeletal Models](https://mlsanity.com/paper/2309.02976/) (2023)
- [Scalable Exploration for High-Dimensional Continuous Control via Value-Guided Flow](https://mlsanity.com/paper/2601.19707/) (2026)
- [Entropy Regularizing Activation: Boosting Continuous Control, Large Language Models, and Image Classification with Activation as Entropy Constraints](https://mlsanity.com/paper/2510.08549/) (2025)

## Built on by (most cited first, 1 of 1)

- [FlashNav: Training Deployable Robot Navigation Policies in Seconds](https://mlsanity.com/paper/2606.15846/) (2026)

## Abstract (arXiv)

Scaling Maximum Entropy Reinforcement Learning (RL) to high-dimensional humanoid control remains a fundamental challenge, as the ''curse of dimensionality'' induces severe exploration inefficiency and training instability. Consequently, highly optimized deterministic policy gradients currently dominate high-throughput regimes. We address this limitation with FastDSAC, a framework that effectively unlocks the potential of maximum entropy stochastic policies for complex continuous control. We introduce Dimension-wise Entropy Modulation (DEM) to dynamically redistribute the exploration budget, alongside a continuous distributional critic tailored to ensure accurate value estimation by mitigating both high-dimensional overestimation and discrete quantization artifacts. Extensive evaluations on HumanoidBench and a diverse set of continuous control tasks demonstrate that FastDSAC establishes state-of-the-art performance for high-dimensional stochastic policies on the evaluated benchmarks. Our method is competitive with and often outperforms strong deterministic baselines, with gains of 180% and 350% on the challenging Basketball and Balance Hard tasks, respectively.