Summary of Restless Bandit Problem with Rewards Generated by a Linear Gaussian Dynamical System, By Jonathan Gornet et al.
Restless Bandit Problem with Rewards Generated by a Linear Gaussian Dynamical System
by Jonathan Gornet, Bruno Sinopoli
First submitted to arxiv on: 15 May 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG); Systems and Control (eess.SY)
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 This paper tackles the fundamental problem of decision-making under uncertainty, which can be formulated as a stochastic multi-armed bandit problem. The learner interacts with an environment by choosing actions at each round, receiving rewards sampled from a stochastic process. The goal is to maximize cumulative reward. Building on linear Gaussian dynamical systems, the authors propose a method for predicting rewards using a modified Kalman filter. They show that previously observed rewards can be used to predict future rewards, regardless of the sequence of previous actions chosen. Numerical evaluations are conducted on linear Gaussian dynamical systems and compared with two well-known stochastic multi-armed bandit algorithms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us make better decisions when things are uncertain. It’s like playing a game where you choose an action and get a reward, but the reward is random. The goal is to make good choices so you get more rewards overall. The authors came up with a new way to predict what rewards you’ll get in the future by combining previous rewards. They tested it on some special kinds of systems and compared it to other popular methods. |
