Summary of Optimization-driven Adaptive Experimentation, by Ethan Che et al.
Optimization-Driven Adaptive Experimentation
by Ethan Che, Daniel R. Jiang, Hongseok Namkoong, Jimmy Wang
First submitted to arxiv on: 8 Aug 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 authors present a novel approach to adaptive experimentation in the face of real-world challenges such as batched and delayed feedback, non-stationarity, multiple objectives and constraints, and personalization. By formulating a dynamic program based on central limit approximations, they enable the use of scalable optimization methods that can incorporate various statistical procedures. The framework is evaluated through a simple heuristic planning method (“solver”) and benchmarked across hundreds of problem instances, demonstrating consistent gains over static randomized control trials. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers designed an adaptive experiment framework to tackle real-world issues in experiments. They developed a mathematical program that can handle many types of objectives, constraints, and statistical methods. This allows for efficient planning using auto-differentiation and GPU parallelization. To test their approach, they used a simple planning method and ran it on hundreds of scenarios with different challenges. The results showed that this framework is better than traditional methods. |
Keywords
» Artificial intelligence » Optimization
