CS271 ยป

CS271 Syllabus

Part I : Fundamentals of Artificial Intelligence (AI)

  • Lesson 1: Welcome to AI (1-2 hours)
    13 clips (28:48), 7 Quizzes

    • Basics and Applications of AI
    • 4 Key Attributes of Environment
      examples: Chess, Poker, Robot cars
    • Sources of Uncertainty
  • Lesson 2: Problem Solving (3-4 hours)
    38 clips (1:01:52), 20 Quizzes

    • Definition of a Problem
      example of Route Finding: Frontier, Region classification, Path Cost
    • Search Types
      Tree search, Graph search
      Breadth-First Search, Depth-First Search, Uniform Cost Search, A* Search
    • Concepts: State spaces, Admissible Heuristic function
  • Problem set 1 (0.5-1 hour) 7 problems

  • Lesson 3: Probability in AI (3-4 hours)
    38 clips (1:03:33), 29 Quizzes

    • Probability: Joint Probability, Conditional Probability, Bayes Rule
    • Bayes Network
    • Independence/Conditional Independence, Explain Away effect, D-separation
  • Lesson 4: Probabilistic Inference (2-3 hours)
    21 clips (48:20), 12 Quizzes

    • Concept: Evidence, Hidden, Query Variables
    • Enumeration vs. Variable Elimination
    • Approximate Inference Methods: Rejection Sampling, Likelihood Weighting, Gibbs Sampling
  • Problem set 2 (0.5-1 hour) 6 problems

  • Lesson 5: Machine Learning (3-4 hours)
    39 clips (1:30:10), 22 Quizzes

    • Machine Learning Taxonomy
    • Maximum Likelihood, Bayes Networks
    • Laplacian Smoothing, Cross-Validation set as overfitting prevention
    • Linear Methods: Classification vs Regression, Gradient descent, Support Vector Machines, K-Nearest Neighbor Algorithm
  • Lesson 6: Unsupervised Learning (2-3 hours)
    30 clips (55:06), 16 Quizzes

    • Clustering: K Means, Expectation Maximization, Gaussian Learning, Spectral Clustering
    • Dimensionality Reduction: Linear Dimensionality Reduction, Piecewise Linear Projection
  • Problem set 3 (1-2 hour) 8 problems

  • Lesson 7: Representation with Logic (2-3 hours)
    13 clips (38:57), 6 Quizzes

    • Propositional Logic: Truth Table, Valid/Satisfiable/Unsatisfiable Sentences
    • First-Order Logic: Structured type representation (c.f. atomic, factored type representation)
  • Lesson 8: Planning (3-4 hours)
    22 clips (1:00:01), 3 Quizzes

    • Problem Solving, Planning, Execution
      Deterministic vs. Stochastic Environment
    • Classical Planning Search Types for Planning: search of the states vs search of the space
    • Situation Calculus
  • Problem set 4 (1-2 hour) 9 problems

  • Lesson 9: Planning under Uncertainty (2-3 hours)
    27 Clips (44:12), 10 Quizzes

    • Markov Decision Process: Policy, Value Functions, Value Iteration
    • Partially Observable Markov Decision Process
  • Lesson 10: Reinforcement Learning (2-3 hours)
    22 Nodes (42:41), 3 Quizzes

    • Forms of Learning: Supervised vs. Unsupervised vs. Reinforcement
    • Reinforcement Learnings
      3 Agents: Utility-Based, Q-Learning, Reflex
      Passive vs Active Agents
  • Problem set 5 (1-2 hour) 3 problems

  • Lesson 11: HMMs and Filters (2-3 hours)
    22 Clips (52:49), 6 Quizzes

    • Hidden Markov Models (HMMs): Markov Chain, Stationary Distribution Example: Localization
    • Particle Filters: Algorithm, pros and cons
      Example: Localization
  • Lesson 12: MDP Review (0.5-1 hours)
    4 Nodes (3:51), 4 Quizzes

  • Midterm (2-3.5 hour) 15 problems

  • Lesson 13: Games (1-2 hours)
    23 Clips (37:50), 13 Quizzes

    • Nature of Games:Stochastic, Partially Observable, Unknown environment, Adversary
      Examples: single-player, two-player games
    • Complexity: Time complexity, Space complexity, Complexity reduction methods and benefits
  • Lesson 14: Game Theory (2-3 hours)
    19 Clips (43:08) / 10 Quizzes

    • Dominant Strategy, Pareto Optimal, Equilibrium
      Example: Game console, Finger Morra
    • Mixed Strategy
    • Mechanism Design
  • Problem set 6 (1-2 hour) 10 problems

  • Lesson 15: Advanced Planning (1-2 hours)
    12 Clips (19:26), 6 Quizzes

    • Scheduling : Time Planning
    • Resources Planning
    • Hierarchical Planning: Hierarchical Task Network,Refinement Planning

Part II : Applications of Artificial Intelligence

  • Lesson 16: Computer Vision I (2-3 hours)
    32 Clips (42:25), 14 Quizzes

    • Image Formation: lens equation
    • Invariance: scale, illumination, rotation, deformation, occlusion, view point
    • Linear Filter:Horizontal, Vertical, Gradient Image filters, Canny Edge Detector, Gaussian Kernel
  • Lesson 17: Computer Vision II (1-2 hours)
    32 Clips (33:42), 10 Quizzes

    • 3D Vision: Depth determination with Stereo
    • Correspondence: Disparity Map, Alignment, Dynamic Programming
  • Lesson 18: Computer Vision III (0.5-1 hours)
    6 Clips (12:33), 3 Quizzes

    • Structure from Motion Models: Projection
    • Mathematical expression: nonlinear Least Square problem
  • Problem set 7 (0.5-2 hour) 6 problems

  • Lesson 19: Robotics I (1-2 hours)
    13 Clips (23:17), 8 Quizzes

    • Example: Autonomous Vehicles
    • Kinematic States, Dynamic States
    • Monte Carlo Localization
  • Lesson 20: Robotics II (1-2 hours)
    13 Clips (30:56), 8 Quizzes

    • Perception: Particle Filter Techniques
      Prediction: Monte Carlo Localization
      Measurement
    • Dynamic Programming: Path Planning, Cost Evaluation
  • Problem set 8 (1-2 hour) 7 problems

  • Lesson 21: Natural Language Processing I (2-3 hours)
    30 Clips (54:26), 10 Quizzes

    • Language Model Type I: Probabilistic Model Markov assumption, Stationary assumption n-gram model Segmentation model Spelling Correction
  • Lesson 22: Natural Language Processing II (2-3 hours)
    12 Clips (42:14), 3 Quizzes

    • Language Model Type II: Logical Model
    • Sentence Structure: Grammar, Parsing, Lexicalized Probabilistic Context Free Grammar
    • Machine Translation: Multi-layer procedure : word-phrase-meaning
      Segmentation, Translation, Distortion models
  • Final Assessment: (2-3 hours) 21 problems

  • Office Hours