Last updated: 3 March 2019

Lecture notes for CM30229 & CM50230
Intelligent Control & Cognitive Systems   2019

Joanna J. Bryson

Dr. Wenbin Li

Lecture 1:  Course Introduction, Intelligence Broadly, & Sensing

Note:  You should have free access to the papers linked below so long as you are on campus or tunneled to campus (e.g. VPN.)

Coursework 1:  Wall Following

Lecture 2:  An Introduction to Artificial Intelligence (its history) & Cognition

Lecture 3:  Action Selection

Lecture 4:  Cognitive Architectures

Lecture 5:  Perception

Lecture 6:  Learning, Neural Networks, and Evolutionary Algorithms.

Lecture 7:  Design & Learnability

Lecture 8:  Science, Agents and Spatial Simulations

Coursework 4:  The Workshop Paper

Postgraduates only (undergraduates get an exam).  The assignment is straight-forward. You should extend one of the first three courseworks to be a conference paper. Please talk to one of the lecturers (we will set up office hours where we can talk individually in person or by video chat during the consolidation week) about what project you want to do. But the paper should be about 2-3 pages of double-column length such as is used by IJCAIAAAI, ACM or IEEE for all of their conferences. The deadline is 9 AM 6 May 2019. To work to distinction, you should actually find some recent workshop or conference papers and use these to establish the current state of the art / knowledge boundary, and then see if you can replicate and / or extend one or more of them.  To work to passing level, you should extend your coursework with more citations to the literature, add more methods and results and possibly another hypothesis for more than one experiment. It should be on a related topic, so share a single motivation & literature review. Each experiment should be objectively evaluated in the results section and its implications discussed in the discussion. This coursework must be conducted individually, but up to 40% of the text may come from your prior co-authored coursework. Note: as of 3 March this coursework is temporary and has not yet been checked.

Tutorial 1:  NetLogo

This lecture is just a howto reviewing code and the NetLogo IDE.  NetLogo is very well documented and supported on line, but you are free to use any ABM environment you choose for CW2, including building your own.  But you should provide links to whatever you use (or all the code if you build your own) so we can double check your code runs.  Here are my lecture notes (to myself):

  1. Show how to download NetLogo
  2. Show how the library / reference & tutorial stuff works
    1. ESPECIALLY the behaviour space tutorial.
  3. Show the three main panes at high level
  4. Show how to add widgets
  5. Run the model, show them the parameter settings.
    1. Promise to talk about the model in second half; dig up that talk.
  6. Show the code.
    1. point out the widgets define globals not in the code, how I deal with that.
    2. in my code I comment these as defs.
  7. Show how monitors & plots work.
  8. Talk about gotchas in synatx, writing own simulators, other platforms (esp. RePast)

Lecture 9:  Social Simulation and Social Structure

Lecture 10:  Hypothesis Testing and Evidence

notes below this line are from 2017 and have not been updated yet for this year.

Coursework 2:  Simulations

Lecture 11: Multiple Conflicting Goals – Intro to Game AI

Lecture 12:  Chatbots, Turing Tests & Believability

Tutorial 2:  BUNG

This lecture is just a howto reviewing BUNG code and the ABOD3 IDE. 

Coursework 3:  Ethical Decisions in Game AI

Lecture 13–14: Culture, Language & Cognition (double-length two-day lecture)

Lecture 15:  Emotions, Drives & Complex Control

Lecture 16:  Consciousness & Cognitive Systems

Lecture 17:  Ethics & Cognitive Systems

Lecture 18:  Regulation & Governance of AI