(Bayesian) Affect Control Theory Lectures

The following lectures were recorded as part of the CS886 (Affective Computing) graduate class at the David R. Cheriton School of Computer Science at the University of Waterloo. They are free to use for non-commercial purposes. Please include a link to this page if you re-distribute use these videos. Information about Bayesact research, including publications and code, can be found at the Bayesact homepage. More details and papers on Affect Control Theory can be found here. Comments can be mailed to jhoey *AT* cs.uwaterloo.ca.

Current lecture: 1 (Social-Cultural Basis of Emotion)

Loading the player ...

Load the stream you want by clicking on the button at the left,
or download the file by clicking on the link at the right.

lecture stream	Date	Duration	Speaker	topic	Download mp4	Slides (PDF)
	Jan 28th, 2014	80 mins	Tobias Schroeder	Social-Cultural Basis of Emotion	(1Gb mp4)	(6.6Mb PDF)
	Jan 30th, 2014	80 mins	Tobias Schroeder	Affect Control Theory	(1.2Gb mp4)	(2.3Mb PDF)
	Feb 4th, 2014	80 mins	Jesse Hoey	Bayesact (I - Background) (you can skip this if you know POMDPs)	(1.1Gb mp4)	(1.1Mb PDF)
	Feb 6th, 2014	80 mins	Jesse Hoey	Bayesact (II - Theory) (also includes some background at the start)	(1.2Gb mp4)	(0.8Mb PDF) (some slides from the previous lecture at the start)
	Feb 11th, 2014	80 mins	Jesse Hoey	Bayesact (III - Applications and results) (includes some previous lecture slides at start)	(1Gb mp4)	(15Mb PDF) (some slides from the previous lecture at the start)

The following screen casts give an introduction to interact and to Bayesact simulations. You can also see these on YouTube

lecture stream	Date	Duration	Speaker	Download mp4	Description
	Dec 2013	7:59 mins	Jesse Hoey	(60Mb mp4)	This screencast shows a basic simulation of a 'tutor' and 'student' in Bayesact and gives an overview of what the output is.
	Dec 2013	16:20 mins	Jesse Hoey	(40Mb mp4)	This screencast shows an example of using the interact java applet alongside the bayesact python simulator. The bayesact simulator is set up in such a way as to mimic as closely as possible the computations of interact. As bayesact doesn't take any shortcuts or make approximations, this requires using a large number of samples (10,000) and have a very small observation noise. As well, the first 5 minutes of this video shows how to set up a basic simulation in interact.
	Dec 2013	5:34 mins	Jesse Hoey	(15Mb mp4)	Simulation of a Bayesact agent with affective identity of "tutor" interacting with a "student", but the bayesact agent does not know this affective identity to start with. Through interactions with the student, the bayesact "tutor" learns that this agent is something like a "student. Interact is used to simulate the actions of the student. It takes bayesact only 2 iterations to figure out the student's identity, as these two identities are fairly close.
	Dec 2013	7:11 mins	Jesse Hoey	(60Mb mp4)	Simulation of a bayesact agent with identity "salesman" interacting with another agent (the "client") who is a "robber", but the bayesact agent does not know this. Through interactions with the robber, the bayesact "salesman" learns that this agent is something like a "robber". Interact is used to simulate the actions of the "robber". It takes about 8 iterations for the bayesact agent to figure this one out, as the two identities are fairly dissimilar (will normally result in high deflection interactions).