Summary of Macrohft: Memory Augmented Context-aware Reinforcement Learning on High Frequency Trading, by Chuqiao Zong et al.
MacroHFT: Memory Augmented Context-aware Reinforcement Learning On High Frequency Trading
by Chuqiao Zong, Chaojie Wang, Molei Qin, Lei Feng, Xinrun Wang, Bo An
First submitted to arxiv on: 20 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Trading and Market Microstructure (q-fin.TR)
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 approach to high-frequency trading (HFT) is proposed, combining reinforcement learning (RL) with hierarchical reinforcement learning (HRL). The method, called MacroHFT, addresses two key issues: overfitting and one-sided decision-making. It consists of two training phases: first, multiple sub-agents are trained on decomposed market data using conditional adapters to adjust their trading policies based on market conditions; second, a hyper-agent is trained to combine the decisions from these sub-agents and output a profitable meta-policy. This approach achieves state-of-the-art performance on minute-level trading tasks in various cryptocurrency markets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary High-frequency trading (HFT) uses special computers to make quick trades in stocks or cryptocurrencies. A new way of doing HFT is proposed, using “reinforcement learning” (RL). RL helps the computer learn from its mistakes and make better decisions over time. The new method, called MacroHFT, fixes two problems with current methods: it avoids making too many trades based on one idea, and it can adjust to changes in the market. This approach shows great results when tested on different cryptocurrency markets. |
Keywords
» Artificial intelligence » Overfitting » Reinforcement learning
