Zhipeng Liang

About

Zhipeng Liang is a researcher specializing in the intersection of deep reinforcement learning and quantitative finance, with a particular focus on algorithmic portfolio management. His work pioneers the application of advanced continuous reinforcement learning algorithms — traditionally developed for game playing and robotic control — to the complex, real-world domain of financial portfolio optimization. Liang's most recognized contribution is his exploration of state-of-the-art policy optimization methods, including Deep Deterministic Policy Gradient (DDPG), Proximal Policy Optimization (PPO), and Policy Gradient (PG), adapted specifically for dynamic portfolio management tasks. His 2018 paper on adversarial deep reinforcement learning in this domain has garnered 57 citations, while a companion work introducing DDPG and PPO frameworks accumulated an additional 23 citations — demonstrating meaningful influence within both the machine learning and computational finance communities. What makes Liang's contributions particularly compelling is his ability to bridge two rapidly evolving fields, offering researchers and practitioners novel frameworks for automating investment decision-making. His work has helped establish deep reinforcement learning as a credible and promising paradigm in financial engineering, inspiring subsequent researchers to further explore AI-driven approaches to portfolio strategy and asset allocation.

Research Focus

Key Achievements

Top Papers

1. 1
   Adversarial Deep Reinforcement Learning in Portfolio Management

   57 citations · 2018
2. 2
   Deep Reinforcement Learning in Portfolio Management

   23 citations · 2018

Key Collaborators

Contact & Links