
Course Overview
Recent breakthroughs in Artificial Intelligence (AI) technologies, notably deep neural networks and reinforcement learning, were largely inspired by our understanding of biological intelligence and how the human brain works. At the same time, effective AI models such as large language models may also provide insights useful for understanding biological intelligence. This course will explore a wide range of topics in AI, with a particular focus on biologically plausible AI agents, including their construction, macro-economics, and current challenges, as well as how biological principles of intelligence can be implemented in AI agents for human-like cognition.
Course topics may include but are not limited to: reinforcement learning algorithms, artificial neural networks, generative AI models, curriculum learning, representation learning, collaborative frameworks for enhancing agent decision-making, human intelligence, and evolution of nervous systems and behavioral complexity. We will review current state-of-the-art AI models, with specific focus on biological plausibility.
There will be weekly discussions, group reviews, and student presentations of recent papers, aiming to explore the potential advantages, complexities, and short-comings of the AI models presented, and how they relate back to biological, psychological, and evolutionary principles.
For anyone interested in receiving updates on our weekly seminars, please subscribe to our mailing list: https://lists.illinois.edu/lists/info/bai (Subscribe button will be located on the left side).