Summary of Improving Portfolio Optimization Results with Bandit Networks, by Gustavo De Freitas Fonseca et al.
Improving Portfolio Optimization Results with Bandit Networks
by Gustavo de Freitas Fonseca, Lucas Coelho e Silva, Paulo André Lima de Castro
First submitted to arxiv on: 5 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Portfolio Management (q-fin.PM)
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 In this paper, researchers propose innovative Bandit algorithms to tackle real-world challenges in Reinforcement Learning (RL), particularly in non-stationary environments. The Adaptive Discounted Thompson Sampling (ADTS) algorithm is introduced, which leverages relaxed discounting and sliding window mechanisms to adapt to changing reward distributions. This approach is extended to the Portfolio Optimization problem through the Combinatorial Adaptive Discounted Thompson Sampling (CADTS) algorithm, addressing computational challenges in Combinatorial Bandits. The proposed algorithms are evaluated using real financial market data, demonstrating superior performance in dynamic environments. Specifically, the bandit network instances outperform classic portfolio optimization approaches, such as capital asset pricing model, equal weights, risk parity, and Markovitz. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about helping computers learn to make better decisions by themselves. Right now, we have algorithms that work well when things stay the same, but in real life, things change all the time! The researchers in this paper created new algorithms called Bandit algorithms to help them adapt to these changes. They used data from financial markets to test their ideas and found that they could make better decisions than old-fashioned methods. |
Keywords
» Artificial intelligence » Optimization » Reinforcement learning
