Readings for Introduction to Intelligent, Cognitive, and Knowledge-Based Systems

• Lectures 1-2: Intelligence, Knowledge, and the Intelligent Agent
  • Optional reading: Allen Newell's 20-questions speech ("You cannot play 20 questions with nature and win").
  • Optional reading: Summary of above
• Lecture 2-3: Planning and PDDL
  • Action selection, perception, actuation: the key processes of intelligent agents
  • Random walker agent, plan-dispatch agent and their respective limitations
  • Optional: This excellent overview presentation of AI planning by Martin Kalisch and Stefan König.
  • Optiona: A presentation with more formal details of the key AI planning algorithms by Bernhard Beckert.
  • Optional: A tutorial on classical planning from 2015 by Malte Helmert and Gabriele Röger. Great discussion of moden heuristics for state-space forward-search classical planning.
• Lectures 4-5: Behvior Arbitraion (Selection and Fusion)
  • Optional: A nice survey of behavior arbitration methods, by Paulo Pirjanian.
  • Optional: Behavior networks for Continuous Domains using Situation-Dependent Motivationsby Klaus Dorer (IJCAI-99).
  • Optional: A Fine-Grained Alternative to the Subsumption Architectue for Mobile Robot Control, by Rosenblatt and Payton (circa 1995).
• Lectures 7-8: Learning in the Agent
  • Optiona: The classical JAIR survey of reinforcement learning from 1996 by Kaelbling, Litman, and Moore. Outdated by still great.
  • Optional: Overview of Intrinsic Rewards by Oudeyer and Kaplan.
  • Optional: The R-MAX model-based reinforcement learning algorithm by Brafman and Tennenholtz