# Bridging the Gap: Enabling Soft Actor Critic for High Performance Legged Locomotion

arXiv: https://arxiv.org/abs/2605.24975 · HTML: https://mlsanity.com/paper/2605.24975/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 24 May 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, 15 of 15)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic Algorithms and Applications](https://mlsanity.com/paper/1812.05905/) (2018)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Humanoid Parkour Learning](https://mlsanity.com/paper/2406.10759/) (2024)
- [Simplifying Deep Temporal Difference Learning](https://mlsanity.com/paper/2407.04811/) (2024)
- [MuJoCo Playground](https://mlsanity.com/paper/2502.08844/) (2025)
- [Legged Robots that Keep on Learning: Fine-Tuning Locomotion Policies in the Real World](https://mlsanity.com/paper/2110.05457/) (2021)
- [Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics](https://mlsanity.com/paper/2501.10100/) (2025)
- [RSL-RL: A Learning Library for Robotics Research](https://mlsanity.com/paper/2509.10771/) (2025)
- [DextrAH-RGB: Visuomotor Policies to Grasp Anything with Dexterous Hands](https://mlsanity.com/paper/2412.01791/) (2024)
- [FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control](https://mlsanity.com/paper/2505.22642/) (2025)
- [Isaac Lab: A GPU-Accelerated Simulation Framework for Multi-Modal Robot Learning](https://mlsanity.com/paper/2511.04831/) (2025)
- [Uncertainty-Aware Robotic World Model Makes Offline Model-Based Reinforcement Learning Work on Real Robots](https://mlsanity.com/paper/2504.16680/) (2025)
- [Learning Sim-to-Real Humanoid Locomotion in 15 Minutes](https://mlsanity.com/paper/2512.01996/) (2025)
- [What Matters for Simulation to Online Reinforcement Learning on Real Robots](https://mlsanity.com/paper/2602.20220/) (2026)

## Abstract (arXiv)

Proximal Policy Optimization (PPO) has become the de facto standard for training legged robots, thanks to its robustness and scalability in massively parallel simulation environments like IsaacLab. However, its on-policy nature makes it inherently sample-inefficient, preventing its use for continuous adaptation and fine-tuning on real hardware. Soft Actor-Critic (SAC), by contrast, is an off-policy algorithm that can reuse past experience, making it a natural candidate for sim-to-real transfer workflows where the same algorithm can be used both in simulation and for online learning on the real robot. Despite these advantages, SAC has consistently failed to match PPO's empirical performance in massively parallel training settings. This work identifies the root causes of this gap and introduces targeted modifications, covering policy initialization, timeout-aware critic targets, and multi-step return estimation, that enable SAC to train stably at scale. Evaluated across multiple legged robot platforms and diverse locomotion tasks, our approach closes the performance gap with PPO entirely.