On January 21, 2026, the Institute of Smart Systems and Artificial Intelligence (ISSAI) hosted a research talk by Dr. Ainur Zhaikhan (PhD, EPFL, Switzerland) entitled “State Estimation and Exploration in Multi-Agent Reinforcement Learning” at Nazarbayev University.
The event opened with a welcome address by ISSAI Founding Director, Professor Atakan Varol, who emphasized the importance of advancing cutting-edge AI research and fostering active academic exchange within the Nazarbayev University community.
In her presentation, Dr. Zhaikhan shared insights from her recent research on multi-agent reinforcement learning, focusing on state estimation under partial observability and the exploration–exploitation trade-offs in distributed systems. Drawing on concepts from social learning, she demonstrated how efficient learning policies can be achieved without relying on explicit state-counting mechanisms.
The talk attracted strong interest from students and researchers, leading to a lively Q&A session and active discussion, reflecting the growing relevance of multi-agent and distributed learning approaches in modern AI research.
About the speaker:
Dr. Ainur Zhaikhan holds a PhD from École polytechnique fédérale de Lausanne (EPFL), Switzerland, where she conducted research on multi-agent reinforcement learning. She earned her bachelor’s degree from Nazarbayev University and a master’s degree from KAUST, Saudi Arabia. Her research focuses on exploration and global state estimation in distributed learning systems.