Summary of Fast Ucb-type Algorithms For Stochastic Bandits with Heavy and Super Heavy Symmetric Noise, by Yuriy Dorn et al.
Fast UCB-type algorithms for stochastic bandits with heavy and super heavy symmetric noise
by Yuriy Dorn, Aleksandr Katrutsa, Ilgam Latypov, Andrey Pudovikov
First submitted to arxiv on: 10 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Optimization and Control (math.OC); Machine Learning (stat.ML)
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 A new method is proposed for constructing UCB-type algorithms for stochastic multi-armed bandits based on general convex optimization methods with an inexact oracle. The regret bounds corresponding to the convergence rates of the optimization methods are derived. A novel algorithm called Clipped-SGD-UCB is introduced, and both theoretical and empirical results demonstrate that it achieves a regret bound of O(log T√KTlogT) for symmetric noise in the reward, outperforming the general lower bound for bandits with heavy-tails. This improved performance is maintained even when the reward distribution lacks an expectation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers has developed a new way to create algorithms that help make good decisions in uncertain situations. They used a type of math called convex optimization to improve these algorithms, which are called UCB-type algorithms. The new algorithm is called Clipped-SGD-UCB and it does better than what was thought possible for certain types of problems. |
Keywords
* Artificial intelligence * Optimization
