# Ablation Study of How Run Time Assurance Impacts the Training and Performance of Reinforcement Learning Agents

arXiv: https://arxiv.org/abs/2207.04117 · HTML: https://mlsanity.com/paper/2207.04117/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 52 references that are in the map.
- First posted on arXiv 8 Jul 2022.

## Builds on (most cited first, 30 of 31)

- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Asynchronous Methods for Deep Reinforcement Learning](https://mlsanity.com/paper/1602.01783/) (2016)
- [OpenAI Gym](https://mlsanity.com/paper/1606.01540/) (2016)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Deep Reinforcement Learning that Matters](https://mlsanity.com/paper/1709.06560/) (2017)
- [Safe Reinforcement Learning via Shielding](https://mlsanity.com/paper/1708.08611/) (2017)
- [Control Barrier Functions: Theory and Applications](https://mlsanity.com/paper/1903.11199/) (2019)
- [End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks](https://mlsanity.com/paper/1903.08792/) (2019)
- [A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems](https://mlsanity.com/paper/1705.01292/) (2017)
- [Provably Efficient Safe Exploration via Primal-Dual Policy Optimization](https://mlsanity.com/paper/2003.00534/) (2020)
- [Safe Reinforcement Learning in Constrained Markov Decision Processes](https://mlsanity.com/paper/2008.06626/) (2020)
- [A Composable Specification Language for Reinforcement Learning Tasks](https://mlsanity.com/paper/2008.09293/) (2020)
- [Safe Reinforcement Learning Using Advantage-Based Intervention](https://mlsanity.com/paper/2106.09110/) (2021)
- [Verifiably Safe Exploration for End-to-End Reinforcement Learning](https://mlsanity.com/paper/2007.01223/) (2020)
- [Constrained Markov Decision Processes via Backward Value Functions](https://mlsanity.com/paper/2008.11811/) (2020)
- [Learning with Safety Constraints: Sample Complexity of Reinforcement Learning for Constrained MDPs](https://mlsanity.com/paper/2008.00311/) (2020)
- [Neural Simplex Architecture](https://mlsanity.com/paper/1908.00528/) (2019)
- [SOTER: A Runtime Assurance Framework for Programming Safe Robotics Systems](https://mlsanity.com/paper/1808.07921/) (2018)
- [Simulation to Scaled City: Zero-Shot Policy Transfer for Traffic Control via Autonomous Vehicles](https://mlsanity.com/paper/1812.06120/) (2018)
- [Neural Network Repair with Reachability Analysis](https://mlsanity.com/paper/2108.04214/) (2021)
- [Learning Safe Neural Network Controllers with Barrier Certificates](https://mlsanity.com/paper/2009.09826/) (2020)
- [Safe Reinforcement Learning Using Robust Action Governor](https://mlsanity.com/paper/2102.10643/) (2021)
- [Verifiable and Compositional Reinforcement Learning Systems](https://mlsanity.com/paper/2106.05864/) (2021)
- [Shield Synthesis for Real: Enforcing Safety in Cyber-Physical Systems](https://mlsanity.com/paper/1908.05402/) (2019)
- [Falsification-Based Robust Adversarial Reinforcement Learning](https://mlsanity.com/paper/2007.00691/) (2020)
- [Scalable Synthesis of Verified Controllers in Deep Reinforcement Learning](https://mlsanity.com/paper/2104.10219/) (2021)
- [Comparing Run Time Assurance Approaches for Safe Spacecraft Docking](https://mlsanity.com/paper/2110.00447/) (2021)

## Built on by (most cited first, 5 of 5)

- [Demonstrating Reinforcement Learning and Run Time Assurance for Spacecraft Inspection Using Unmanned Aerial Vehicles](https://mlsanity.com/paper/2405.06770/) (2024)
- [Run Time Assured Reinforcement Learning for Six Degree-of-Freedom Spacecraft Inspection](https://mlsanity.com/paper/2406.11795/) (2024)
- [Space Processor Computation Time Analysis for Reinforcement Learning and Run Time Assurance Control Policies](https://mlsanity.com/paper/2405.06771/) (2024)
- [Investigating the Impact of Choice on Deep Reinforcement Learning for Space Controls](https://mlsanity.com/paper/2405.12355/) (2024)
- [Stacked Universal Successor Feature Approximators for Safety in Reinforcement Learning](https://mlsanity.com/paper/2409.04641/) (2024)

## Abstract (arXiv)

Reinforcement Learning (RL) has become an increasingly important research area as the success of machine learning algorithms and methods grows. To combat the safety concerns surrounding the freedom given to RL agents while training, there has been an increase in work concerning Safe Reinforcement Learning (SRL). However, these new and safe methods have been held to less scrutiny than their unsafe counterparts. For instance, comparisons among safe methods often lack fair evaluation across similar initial condition bounds and hyperparameter settings, use poor evaluation metrics, and cherry-pick the best training runs rather than averaging over multiple random seeds. In this work, we conduct an ablation study using evaluation best practices to investigate the impact of run time assurance (RTA), which monitors the system state and intervenes to assure safety, on effective learning. By studying multiple RTA approaches in both on-policy and off-policy RL algorithms, we seek to understand which RTA methods are most effective, whether the agents become dependent on the RTA, and the importance of reward shaping versus safe exploration in RL agent training. Our conclusions shed light on the most promising directions of SRL, and our evaluation methodology lays the groundwork for creating better comparisons in future SRL work.