Optimizing Algorithmic Trading Through DRL: A Comparative Analysis of Single-Agent and Multi-Agent Models
- Title
- Optimizing Algorithmic Trading Through DRL: A Comparative Analysis of Single-Agent and Multi-Agent Models
- Creator
- Mani Shankar, M.; Sweety, A.; Deepthi, Das
- Description
- This work investigates how Deep Reinforcement Learning (DRL) can elevate algorithmic tradingespecially in fast-paced, high-frequency markets. We propose a full-fledged framework to compare different setups, from solo agents to multi-agent systems, applying DRL methods like Proximal Policy Optimization (PPO), Deep Q-Network (DQN), and Advantage Actor-Critic (A2C), along with combinations of these. We trained on hourly stock data from 24 firms over two years (Jan 1, 2020Jan 1, 2022) and tested performance over the next year (Jan 1, 2022Jan 1, 2023). We evaluated key factorsreturns, risk control, and how well these models adapt to changing markets. The single-agent PPO model stood out, achieving a remarkable profit factor of 28.07 on BIDU and keeping peak drawdowns frequently under 1%. This demonstrates both solid capital protection and high risk-adjusted performance. Ensemble models showed balanced performance in both single-agent and multi-agent setups, achieving a Sharpe ratio of 0.75 and Sortino ratios up to 7.7, outperforming existing benchmarks. Comparative analyses revealed that ensemble strategies enhance market responsiveness and improve both stability and profitability in volatile environments. Sensitivity analysis confirmed the robustness of model performance across various hyperparameter settings. Overall, the proposed DRL-based ensemble framework demonstrates strong potential to improve real-world HFT systems by delivering more adaptive, stable, and efficient algorithmic trading solutions. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
- Source
- Lecture Notes in Networks and Systems;Volume;1723 LNNS;pp.1-15
- Date
- 01-01-2026
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Subject
- Algorithmic trading; Deep reinforcement learning; Ensemble models; High-frequency trading; Proximal policy optimization
- Coverage
- Mani Shankar M., Department of Statistics and Data Science, Christ University, Bengaluru, India; Sweety A., Department of Statistics and Data Science, Christ University, Bengaluru, India; Deepthi D., Department of Statistics and Data Science, Christ University, Bengaluru, India
- Rights
- Restricted Access; Hardcopy may be available in the library
- Relation
- ISSN: 23673370; ISBN: 978-303210782-4;
- Format
- online
- Language
- English
- Type
- Conference paper
Collection
Citation
Mani Shankar, M.; Sweety, A.; Deepthi, Das, “Optimizing Algorithmic Trading Through DRL: A Comparative Analysis of Single-Agent and Multi-Agent Models,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 18, 2026, https://archives.christuniversity.in/items/show/25373.