Summary of Beacon: a Bayesian Optimization Strategy For Novelty Search in Expensive Black-box Systems, by Wei-ting Tang and Ankush Chakrabarty and Joel A. Paulson
BEACON: A Bayesian Optimization Strategy for Novelty Search in Expensive Black-Box Systems
by Wei-Ting Tang, Ankush Chakrabarty, Joel A. Paulson
First submitted to arxiv on: 5 Jun 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG)
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 proposed sample-efficient novelty search (NS) method, inspired by Bayesian optimization principles, handles model opacity by modeling the input-to-behavior mapping with multi-output Gaussian processes (MOGP). This approach balances exploration-exploitation trade-offs and is scalable with respect to data and inputs. The algorithm outperforms existing baselines on benchmark problems and real-world examples, finding larger sets of diverse behaviors under limited sampling budgets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to find many different ways for a system to behave is proposed. This “novelty search” can be used in engineering design problems like material discovery or robot navigation. The problem is that some systems are hard to simulate or evaluate, so the algorithm must handle this “model opacity”. To do this, it uses a special kind of math called multi-output Gaussian processes (MOGP). This allows the algorithm to find the best inputs to test that will give new and interesting results while also trying not to waste time on things that have already been tried. The result is an efficient way to find many different behaviors in systems that are hard to work with. |
Keywords
» Artificial intelligence » Optimization
