Summary of Non-stationary Latent Auto-regressive Bandits, by Anna L. Trella et al.
Non-Stationary Latent Auto-Regressive Bandits
by Anna L. Trella, Walter Dempsey, Asim H. Gazi, Ziping Xu, Finale Doshi-Velez, Susan A. Murphy
First submitted to arxiv on: 5 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 novel online linear contextual bandit algorithm called Latent AR LinUCB (LARL) is proposed for the non-stationary multi-armed bandit (MAB) problem. Unlike existing methods, LARL does not rely on a budget for non-stationarity and instead forms good predictions of reward means by implicitly predicting a latent, auto-regressive state. By reducing the problem to a linear dynamical system, LARL approximates a steady-state Kalman filter and efficiently learns system parameters online. The algorithm provides an interpretable regret bound with respect to the level of non-stationarity in the environment, achieving sub-linear regret if the noise variance is sufficiently small. Empirically, LARL outperforms various baseline methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine a situation where you need to make decisions based on changing circumstances. This paper proposes a new way to solve this problem called Latent AR LinUCB (LARL). It’s an algorithm that can learn from changing patterns and make good decisions without knowing the rules of the game ahead of time. LARL works by predicting the underlying patterns in the data and using those predictions to make informed decisions. |
