Summary of Safety Filters For Black-box Dynamical Systems by Learning Discriminating Hyperplanes, By Will Lavanakul et al.
Safety Filters for Black-Box Dynamical Systems by Learning Discriminating Hyperplanes
by Will Lavanakul, Jason J. Choi, Koushil Sreenath, Claire J. Tomlin
First submitted to arxiv on: 7 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach for designing safety filters for black-box dynamical systems using learning-based methods. The authors aim to eliminate dependence on specific certificate functions, such as Control Barrier Functions (CBFs) and Hamilton-Jacobi (HJ) reachability value functions, by defining a discriminating hyperplane that shapes the half-space constraint on control input at each state. This approach generalizes traditional safety methods and simplifies safety filter design. The authors present two strategies for learning the discriminating hyperplane: supervised learning using pre-verified control invariant sets for labeling, and reinforcement learning (RL) without requiring labels. The method separates performance and safety, enabling a reusable safety filter for learning new tasks without retraining from scratch. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper introduces a new way to make sure complex systems stay safe by using machine learning techniques. Instead of relying on specific formulas, the authors create a special line that tells the system what kind of control inputs are allowed at each moment. This approach is more flexible and can be used for many different tasks without needing to start from scratch every time. |
Keywords
* Artificial intelligence * Machine learning * Reinforcement learning * Supervised
