# Locally Persistent Exploration in Continuous Control Tasks with Sparse Rewards

arXiv: https://arxiv.org/abs/2012.13658 · HTML: https://mlsanity.com/paper/2012.13658/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Dec 2020.

## Built on by (most cited first, 13 of 13)

- [A Survey of Exploration Methods in Reinforcement Learning](https://mlsanity.com/paper/2109.00157/) (2021)
- [TAAC: Temporally Abstract Actor-Critic for Continuous Control](https://mlsanity.com/paper/2104.06521/) (2021)
- [Dynamic Decision Frequency with Continuous Options](https://mlsanity.com/paper/2212.04407/) (2022)
- [Generative Planning for Temporally Coordinated Exploration in Reinforcement Learning](https://mlsanity.com/paper/2201.09765/) (2022)
- [Active Sensing with Predictive Coding and Uncertainty Minimization](https://mlsanity.com/paper/2307.00668/) (2023)
- [Deep Exploration with PAC-Bayes](https://mlsanity.com/paper/2402.03055/) (2024)
- [SFP: State-free Priors for Exploration in Off-Policy Reinforcement Learning](https://mlsanity.com/paper/2205.13528/) (2022)
- [Subwords as Skills: Tokenization for Sparse-Reward Reinforcement Learning](https://mlsanity.com/paper/2309.04459/) (2023)
- [Reinforcement Learning with Elastic Time Steps](https://mlsanity.com/paper/2402.14961/) (2024)
- [Simultaneously Updating All Persistence Values in Reinforcement Learning](https://mlsanity.com/paper/2211.11620/) (2022)
- [MOSEAC: Streamlined Variable Time Step Reinforcement Learning](https://mlsanity.com/paper/2406.01521/) (2024)
- [TARC: Time-Adaptive Robotic Control](https://mlsanity.com/paper/2510.23176/) (2025)
- [Variable Time Step Reinforcement Learning for Robotic Applications](https://mlsanity.com/paper/2407.00290/) (2024)

## Abstract (arXiv)

A major challenge in reinforcement learning is the design of exploration strategies, especially for environments with sparse reward structures and continuous state and action spaces. Intuitively, if the reinforcement signal is very scarce, the agent should rely on some form of short-term memory in order to cover its environment efficiently. We propose a new exploration method, based on two intuitions: (1) the choice of the next exploratory action should depend not only on the (Markovian) state of the environment, but also on the agent's trajectory so far, and (2) the agent should utilize a measure of spread in the state space to avoid getting stuck in a small region. Our method leverages concepts often used in statistical physics to provide explanations for the behavior of simplified (polymer) chains in order to generate persistent (locally self-avoiding) trajectories in state space. We discuss the theoretical properties of locally self-avoiding walks and their ability to provide a kind of short-term memory through a decaying temporal correlation within the trajectory. We provide empirical evaluations of our approach in a simulated 2D navigation task, as well as higher-dimensional MuJoCo continuous control locomotion tasks with sparse rewards.