History of AI including recent developments (2 lectures): the early history of AI including the calculus ratiocinator, the Church-Turing Thesis, the significance of the Dartmouth Conference, the development of expert systems, the fifth generation computer project, the AI winter, and the development of Deep Blue. Recent developments will be introduced by discussing, for example, IBM's Watson, AlphaGo, and the DARPA Grand Challange. The examples of recent developments are revisited to motivate the introduction of search problems, reasoning under uncertainty, knowledge representation, and learning in subsequent lectures. Problem-Solving Through Search (8 lectures): Problem formulation; uninformed search strategies; informed search strategies; constraint satisfaction problems; adversarial search. Reasoning under Uncertainty (9 lectures): Probability in AI; axioms of probability; joint distribution; independence; Bayes' rule; Bayesian networks. Knowledge Repre
sentation (4 lectures): Logic; logical agents; knowledge engineering; inference; planning; Goedel's incompleteness theorem. Learning (4 lectures): Different forms of learning; reinforcement learning. Philosophy and ethics of AI (3 lectures): Introduction to the questions 'Can a machine act intelligently?' and 'Can a machine have mental states?'; in particular, the Turing Test and Searle's Chinese room argument are introduced. Ethics of AI is introduced by discussing machine ethics and weaponization of AI.
|