# Discriminative Barrier Functions for Safe Adversarial Imitation Learning from Observation

arXiv: https://arxiv.org/abs/2607.13938 · HTML: https://mlsanity.com/paper/2607.13938/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safe offline RL, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 22 of its 43 references that are in the map.
- First posted on arXiv 15 Jul 2026.

## Placed in

- Offline RL and imitation learning from data / [Safe offline RL](https://mlsanity.com/offlinerl/safe-offline-rl/)
- Deep reinforcement learning methods / [Safe and constrained RL](https://mlsanity.com/deeprl/safe-and-constrained-rl/)

## Builds on (most cited first, 22 of 22)

- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Behavioral Cloning from Observation](https://mlsanity.com/paper/1805.01954/) (2018)
- [Control Barrier Functions: Theory and Applications](https://mlsanity.com/paper/1903.11199/) (2019)
- [Learning Control Barrier Functions from Expert Demonstrations](https://mlsanity.com/paper/2004.03315/) (2020)
- [What Matters for Adversarial Imitation Learning?](https://mlsanity.com/paper/2106.00672/) (2021)
- [How to Train Your Neural Control Barrier Function: Learning Safety Filters for Complex Input-Constrained Systems](https://mlsanity.com/paper/2310.15478/) (2023)
- [Synthesis of Control Barrier Functions Using a Supervised Machine Learning Approach](https://mlsanity.com/paper/2003.04950/) (2020)
- [GCBF+: A Neural Graph Control Barrier Function Framework for Distributed Safe Multi-Agent Control](https://mlsanity.com/paper/2401.14554/) (2024)
- [MuSHR: A Low-Cost, Open-Source Robotic Racecar for Education and Research](https://mlsanity.com/paper/1908.08031/) (2019)
- [CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions](https://mlsanity.com/paper/2510.14959/) (2025)
- [In-Distribution Barrier Functions: Self-Supervised Policy Filters that Avoid Out-of-Distribution States](https://mlsanity.com/paper/2301.12012/) (2023)
- [Learning Risk-Aware Costmaps via Inverse Reinforcement Learning for Off-Road Navigation](https://mlsanity.com/paper/2302.00134/) (2023)
- [Walking in Narrow Spaces: Safety-critical Locomotion Control for Quadrupedal Robots with Duality-based Optimization](https://mlsanity.com/paper/2212.14199/) (2022)
- [Neural Control Barrier Functions for Safe Navigation](https://mlsanity.com/paper/2407.19907/) (2024)
- [How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraints](https://mlsanity.com/paper/2511.18606/) (2025)
- [Learning Control Barrier Functions and their application in Reinforcement Learning: A Survey](https://mlsanity.com/paper/2404.16879/) (2024)
- [V-OCBF: Learning Safety Filters from Offline Data via Value-Guided Offline Control Barrier Functions](https://mlsanity.com/paper/2512.10822/) (2025)
- [Safe Inverse Reinforcement Learning via Control Barrier Function](https://mlsanity.com/paper/2212.02753/) (2022)
- [Risk-Sensitive Inhibitory Control for Safe Reinforcement Learning](https://mlsanity.com/paper/2310.01538/) (2023)
- [Model Predictive Adversarial Imitation Learning for Planning from Observation](https://mlsanity.com/paper/2507.21533/) (2025)
- [Lipschitz Continuity of Signal Temporal Logic Robustness Measures: Synthesizing Control Barrier Functions from One Expert Demonstration](https://mlsanity.com/paper/2304.03849/) (2023)

## Abstract (arXiv)

Inverse Reinforcement Learning (IRL) algorithms are powerful tools for learning from and generalizing expert demonstrations, but they often rely on unconstrained exploration, rendering them unsafe for real-world deployment. Meanwhile, Control Barrier Functions (CBFs) can guarantee the safety of control systems, but the analytical design of CBFs can be time-consuming and esoteric. In this work, we address these limitations jointly by constraining reward function candidacy during IRL to the space of CBFs, yielding a formulation that exhibits safe online control with continuous experiential improvement. Crucially, this framework enables the data-driven recovery of barrier functions directly from unlabeled expert observations. We demonstrate that the recovered barrier function is robust to unsafe states entirely absent from the expert data. Furthermore, we benchmark our method against standard IRL baselines in a simulated navigation environment, demonstrating improved safety performance. Finally, we investigate the trade-offs of planning-based versus policy-based IRL methods across both simulation and a real world obstacle avoidance task.