Skip to main content

Gautam Biswas

CS 6360: Advanced Artificial Intelligence

Course Objectives: Artificial Intelligence spans a wide variety of topics at the forefront of computer science research. This includes areas like machine learning, robotics, planning, computer vision, natural language processing, optimization, and many others. This course will introduce many of these topics, but it will be taught at the graduate level, where students will delve into specific algorithms and applications in significant detail.  In many ways, this course serves as a gateway to more advanced AI and machine learning courses. Students will explore this through problem-solving paradigms, informed search algorithms, local search using evolutionary algorithms (e.g., genetic algorithms), planning and scheduling methods, reasoning with uncertain knowledge, and machine learning methods that support reasoning and problem-solving. Application topics will include computer vision, robot motion planning, and natural language processing.

Text:

Artificial Intelligence A Modern Approach, by Stuart Russell and Peter Norvig, Fourth edition, Prentice Hall, NJ, 2009.

More information is available at http://aima.cs.berkeley.edu.

Course content will include additional material from relevant publications. These publications will be posted on Brightspace.

Topics Covered:

  • Introduction to AI  & Intelligent Agents
  • Search and Problem Solving
    • A*, IDA*, Anytime WA*
  • Adversarial Search (Two person games)
    • Alpha-Beta, Monte Carlo Tree Search (MCTS), & Stochastic Games
  • Local Search and Optimization Problems
    • Simulated Annealing, Evolutionary Search Algorithms (Genetic Algorithms), and Search in Continuous spaces (Gradient Descent Methods)
  • Planning & Scheduling
    • Planning Graphs, Planning Space Planning (PSP), Partial Order Planning (POP), Hierarchical Task Network (HTN) Planning, Scheduling & Resource Allocation
  • Reasoning under Uncertainty
    • Bayesian Networks, Learning Bayesian Networks
  • Reasoning under Uncertainty over Time
    • Kalman Filters, Hidden Markov Models, Dynamic Bayes Nets
  • Decision Making
    • Utility Theory, Decision Networks, Sequential Decision Problems, Markov Decision Processes (MDPs), Partially Observable MDPs (POMDPs)
  • Machine Learning
    • Supervised Learning, Unsupervised Learning, Reinforcement Learning
  • Additional Topics of Interest
    • Deep Learning, Computer Vision, Natural Language Processing