Describing a task as a search problem — initial state, actions, transition model, goal test, and path cost — and abstracting a physical environment into states and a graph.
Uninformed and informed strategies — BFS, DFS, Uniform-Cost, Greedy Best-First, A*, IDS, and SMA* — comparing frontier dynamics, explored sets, node expansion order, and optimality.
Multi-agent competitive environments — game formulation, Minimax value propagation, Alpha-Beta pruning, heuristic evaluation with depth cutoffs, chance games, and Monte Carlo Tree Search (MCTS).
Variables, domains, and constraints — constraint graphs, node and arc consistency (AC-3), backtracking with MRV, degree and LCV heuristics, forward checking and MAC, min-conflicts local search, and tree-structured CSPs.
Representing facts & rules via Knowledge-Based Agents (Propositional Logic, Forward Chaining), reasoning under uncertainty with Bayesian Networks, and dynamic state tracking with Hidden Markov Models.
Supervised learning (BoW, TF-IDF, Naive Bayes, Metrics), Neural Networks (Perceptron, Gradients), Unsupervised K-Means, and Reinforcement Learning (MDPs, Q-Learning).
A goal-driven agent that reasons about a task, plans, acts, observes results, and replans — using tools and memory, and coordinating with other agents to complete work.