A New DAPO Algorithm for Stock Trading
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916756164968448 |
|---|---|
| author | Zha, Ruijian Liu, Bojun |
| author_facet | Zha, Ruijian Liu, Bojun |
| contents | Recent advances in reinforcement learning, such as Dynamic Sampling Policy Optimization (DAPO), show strong performance when paired with large language models (LLMs). Motivated by this success, we ask whether similar gains can be realized in financial trading. We design a trading agent that combines an improved Group Relative Policy Optimization (GRPO) algorithm, augmented with ideas from DAPO, with LLM-based risk and sentiment signals extracted from financial news. On the NASDAQ-100 index (FNSPID dataset), our agent attains a cumulative return of 230.49 percent and an information ratio of 0.37, outperforming the CPPO-DeepSeek baseline. It also cuts training time from about 8 hours to 2.5 hours over 100 epochs while markedly reducing RAM usage. The proposed RL-LLM framework offers a scalable path toward data-efficient trading agents. Code: https://github.com/Ruijian-Zha/FinRL-DAPO-SR/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_06408 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A New DAPO Algorithm for Stock Trading Zha, Ruijian Liu, Bojun Computational Engineering, Finance, and Science 68T05, 91G60 I.2.6; J.4; G.3 Recent advances in reinforcement learning, such as Dynamic Sampling Policy Optimization (DAPO), show strong performance when paired with large language models (LLMs). Motivated by this success, we ask whether similar gains can be realized in financial trading. We design a trading agent that combines an improved Group Relative Policy Optimization (GRPO) algorithm, augmented with ideas from DAPO, with LLM-based risk and sentiment signals extracted from financial news. On the NASDAQ-100 index (FNSPID dataset), our agent attains a cumulative return of 230.49 percent and an information ratio of 0.37, outperforming the CPPO-DeepSeek baseline. It also cuts training time from about 8 hours to 2.5 hours over 100 epochs while markedly reducing RAM usage. The proposed RL-LLM framework offers a scalable path toward data-efficient trading agents. Code: https://github.com/Ruijian-Zha/FinRL-DAPO-SR/ |
| title | A New DAPO Algorithm for Stock Trading |
| topic | Computational Engineering, Finance, and Science 68T05, 91G60 I.2.6; J.4; G.3 |
| url | https://arxiv.org/abs/2505.06408 |