Haiyan Zhao

About

Haiyan Zhao is a leading researcher in self-adaptive systems, with a primary focus on leveraging reinforcement learning and meta-learning to create intelligent, autonomous software. Her major contributions lie in developing frameworks that enable systems to dynamically learn and update their adaptation policies in response to unforeseen environmental changes. In her highly cited 2021 work, "A Meta Reinforcement Learning-based Approach for Self-Adaptive System" (10 citations), she introduced a self-learning adaptive system that uses meta-reinforcement learning to achieve high performance in dynamic environments, setting a new standard for autonomous adaptability. Building on this, her 2022 paper "Goal-oriented Knowledge Reuse via Curriculum Evolution for Reinforcement Learning-based Adaptation" (3 citations) addresses a critical challenge in the field: efficiently reusing knowledge to speed up policy relearning during adaptation. By proposing a curriculum evolution approach, Zhao enables systems to transfer learned knowledge across different goals, significantly improving learning efficiency. Her work is foundational for the next generation of self-adaptive systems, with applications in robotics, cloud computing, and autonomous vehicles. Zhao’s research is shaping how machines learn to adapt, making her a key figure in the evolution of intelligent, resilient software systems.

Research Focus

Key Achievements

Top Papers

1. 1
   A Meta Reinforcement Learning-based Approach for Self-Adaptive System

   10 citations · 2021
2. 2
   Goal-oriented Knowledge Reuse via Curriculum Evolution for Reinforcement Learning-based Adaptation

   3 citations · 2022

Key Collaborators

Contact & Links