Introduction to Artificial Intelligence

What you'll learn

• Explain the main definitions of AI, describe problems as rational agents in environments, and distinguish AI, machine learning, deep learning and generative AI
• Trace the history of AI from the Dartmouth workshop through the AI winters to deep learning and large language models, and draw lessons for judging current claims
• Formulate problems as state-space search, compare blind search with A* and heuristics, and explain minimax, game-playing systems and optimization
• Reason with logic, rules and knowledge graphs, apply Bayes' rule with base rates, and choose actions by expected utility
• Describe supervised, unsupervised and reinforcement learning, fit a model by gradient descent, recognize overfitting, and evaluate classifiers with precision and recall
• Explain how deep neural networks learn through backpropagation, compare network types, and recognize overfitting and the limits of deep learning
• Apply clustering, dimensionality reduction and anomaly detection, explain self-supervised learning, and describe reinforcement learning with Q-learning and reward hacking
• Explain how AI represents and interprets images, sound and text, including convolution, spectrograms and embeddings, and where perception fails
• Explain how large language models are built and deployed, estimate their compute, and identify their main failure modes and how to work around them
• Plan an applied AI project from problem framing and data to evaluation, deployment, drift monitoring and governance
• Analyze fairness, privacy, misinformation, work, safety and environmental issues raised by AI, and describe how governments regulate it and what responsible practice requires
• Describe AI agents, robotics, AI for science, scaling trends and debates about advanced AI, and plan how to build AI skills