An original inspiration of AI, from the early days of McCulloch and Pitts, is the human brain. The aim of this module is not a full Neuroscience or Cognitive Science module, but rather one covering topics that have inspired, or is in the process of inspiring AI research. These range from microscopic to macroscopic to structural: Hebbian learning, simple perceptrons, the canonical microcircuit, predictive coding (as a cognitive model and AI parallel), and dopamine-coded reward prediction for Reinforcement Learning. On a larger scale: hippocampal feedback, including grid and place cells. Related to convolutional networks, the module will cover the visual hierarchy. Related to recurrent networks, the module will present an overview of the auditory processing system. The module will conclude with open problems in motor control (neuroscientifically and AI) which are not well characterised, and harder to formulate in current AI frameworks.
Upon completion of the module the student will be able to:
- give a broad account of the history of brain-inspired AI;
- summarise and discuss basic elements of Neuro- and Cognitive Science, particularly brain structure and organisation in AI, from the point of view of mathematical modelling and algorithms;
- formulate and interpret the theory of visual pathway and memory from an AI perspective;
- discuss and appraise open problems in brain-inspired AI.