Tea-time talks

	Reinforcement Learning and Artificial Intelligence (RLAI)
	Tea-time Talks

The RLAI group at the University of Alberta is having a tea-time talk every Monday to Thursday this summer at 4:00-4:30pm in CSC 333. Refreshments will be provided, starting at 3:45pm. Starting September 6, tea-time talks will be held every Wednesday at 4:30-5pm in CSC 333 with tea and cookies available at 4:15pm.

The intention of the tea-time talks is to efficiently transmit information on a variety of current Reinforcement Learning topics.

The ambition of this page is to organise the tea-time talks and provide a mailing list for all participants.

Guidelines

A tea-time talk can either be an overview of a relevant paper or a summary of the presenter's own work
Talks start punctually at 4:00pm, and are strictly limited to 20 minutes + 10 minutes of question time
Presenters should focus on transmitting information as efficiently as possible
It is the presenter's responsibility to pass over disruptive questions if they divert from the talk
It is the listener's responsibility to only ask questions that help clarify rather than disrupt the talk
All subscribers to this list are expected to give one short talk per round, no "hangers on" please!
Short of ideas? Check out the paper suggestions below

Organisation

These web pages are "self-organising"
For each new round of talks, all subscribers will be randomly assigned a day to present a talk
Please resolve date conflicts by agreeing to swap with someone else, and then edit the table below
If you have difficult constraints, notify the list to find someone to swap with (click on "notify" at foot of page)
Please select a topic for your talk at least one week in advance, and edit it into the table below
If you are presenting a paper, please provide a link to the paper in the table
Reminders of upcoming talks will be sent out on a weekly basis
Editing open pages such as this is easy but requires Netscape or Mozilla or this Python script (instructions here). You really should learn how to do this, but don't let it slow down getting your information in here. If you don't know how to edit yet, then send your information by email to someone who can put it in, such as Anna or David
One student per week is in charge of picking up tea and cookies from Lori, making sure the kettle is boiled by 3:45, and clearing away at the end of the talk. This responsibility will rotate through the students in order of their presentations, and is marked in square brackets below.

Round 1: (May 15 - July 20)
Round 2: current

Round 1: (May 15 - July 20)

Date	Presenter	Topic	Link
May 15-18
Mon	Rich Sutton	RL in finance paper by Kearns et alia	NFK06
Tue	Michael Bowling	Reduced Rank Regression
Wed	Dan Lizotte	Intrinsic Reward Exploration (Simsek & Barto)
Thu	Dale Schuurmans	Boosting
May 22-25	[Organiser: Nolan]
Mon		[Victoria Day]
Tue	Alborz	RL in Matrix Form	TD0GD
Wed	Mark Ring	Konidaris, Barto: Knowledge Transfer in RL (ICML 2006)	autoshape
Thu	James Neufeld	Image Features Using SIFT	SIFT
May 29-June 1	[Organiser: Brad]
Mon	Amir massoud Farahmand	Policy Gradient in Continuous Time (Remi Munos)	JMLR06
Tue	Russ Greiner	KDD Cup 2006	KDDCup
Wed	Anna Koop	Information Theoretic Manifold Discovery	PDF
Thu		[Cancelled]
June 12-15	[Organiser: Alborz]	[Note: 1 week break for NIPS]
Mon	nolan	Constructing Informative Priors using Transfer Learning	.pdf
Tue		[Cancelled for Ken Thompson]
Wed	masoud	RL with Context Detection (ICML 2006)	PDF
Thu	bjjoyce	Using Inaccurate Models in RL (ICML 2006)	PDF
June 19-22	[Organiser: James]
Mon	Tao Wang	An Analytic Solution to Discrete Bayesian RL	PDF
Tue	maz
Wed	johanson
Thu	coulthar	Graph Visualization using Disc Trees
July 3-6	[Organiser: Anna]	[1 week break for IJCAI deadline and ICML conference]
Mon		[Canada Day]
Tue	Scott Sanner	Relational RL and feature discovery	SS06
Wed	Vadim Bulitko	AIIDE report
Thu	carmelo	Hierarchical Temporal Memory (Hawkins, George, Numenta Inc.)	PDF
July 10-13	[Organiser: Brian]
Mon	armita
Tue	yaki	Variance of discounted Markov decision processes (Matthew Sobel)	JSTOR82
Wed	erafols	Representing Systems with Hidden State (AAAI 2006) (C. Hundt, P. Panagaden, J. Pineau, D. Precup)	PDF
Thu	cosmin	Planning with Approximate, Learned Models
July 17-20	[Organiser: Masoud]
Mon	awhite [CSC 3-49]	Cognitive Maps in Rats and Robots (Vernena Hafner)
Tue	silver [CSC 3-49]	Monte-Carlo Tree Search (Remi Coulom)	RC06
Wed	mlee	Experience-Efficient Learning in Associative Bandit Problems	PDF
Thu	marthal	Predictive State Representations with Options (Wolfe and Satinder)

Round 2: (July 24- )

Date	Presenter	Topic	Link
July 24 (Monday)		Second round of TTT: organizational meeting
July 31-August 3	[Organiser: Sergey]
Mon	Alborz Geramifard	AAAI 2006 and AIIDE 2006 in a glance
Tue	Sergey Kirshner	Trees: what are they good for?	ChowLiu68
Wed	Brian Tanner	The Wumpus World as a challenge problem for representations and reinforcement learning
Thu	Anna Koop	CogSci 2006 summary	CogSci2006
August 7-10	[Organiser: Dan]
Mon		[Heritage Day]
Tue	Martin Zinkevich	Teaching and Learning
Wed	David Silver	Planning, predictive representations and two-player games
Thu	James Kehoe	Real-Time Processes are Paramount in Classical and Operant Conditioning
August 14-17	[Organiser: Tao]
Mon	Elliot Ludvig	Why Rats are Smarter than Pigeons: Lessons for AI from Animal Timing
Tue	Mohammad Ghavamzadeh	On the Nystrom Method for Approximating a Gram Matrix for Improved Kernel-Based Learning	PDF
Wed	Rich Sutton	the TD model of classical conditioning	pdf
Thu	Michael Bowling	If it worked for NLP...
August 21-24	[Organiser: Michael]
Mon	Brian Tanner	Predictive Action Descriptions from Experience (please come I need help)
Tue	Amir massoud Farahmand	Samuel Meets Amarel: Automating Value Function Approximation using Global State Space Analysis	AAAI05
Wed	Martin Zinkevich	Maximum Margin Planning	PDF
Thu	Mark Lee	Does the Turing Test Demonstrate Intelligence or Not?	PDF
August 28-31	[Organiser: Cosmin]
Mon	Rob Holte	Coarse-to-Fine Dynamic Programming
Tue	Carmelo Piccione	How fast to work:Response vigor, motivation, and tonic dopamine (Niv et al, NIPS 2005)	PDF
Wed	James Neufeld [CSC 3-49]	A brief overview of predictive linear Gaussian (PLG) models	PDF
Thu		[Cancelled]

Fall Schedule		Talks every Wednesday, 4:30-5pm in CSC 3-33
September 6	Cameron Upright	A 3-dimensional simulation of the Octopus arm
September 13		[Cancelled]
September 20	James Kehoe	Is the generation of a learned response linear or catastrophic?
September 27	Katherine Davison [CSC 3-49]	Kernel Predictive Linear Gaussian Models
October 4	Vadim Bulitko [CSC 3-49]	Learning target model in moving target pursuit	Slides
October 11		[Cancelled]
October 18	Mark Ring / Tao Wang	The Simulated Adaptive Behavior (SAB) Conference, held in Rome, September 2006	SAB'06
October 25		[Cancelled]
November 1	Csaba Szepesvari	Exploration-Exploitation Algorithms for finite MDPs	BK97
November 8	Nolan Bard	An Introduction to Bayesian Filtering Techniques
November 15	Mohammad Ghavamzadeh	Linearly-Solvable Markov Decision Problems	PDF
November 22		[Cancelled]
November 29	Eric Coulthard	Netflix Million Dollar Prize	Link
December 6		[NIPS]
December 13
Future
	johanson
	Andrew Butcher	Temporal Difference Models and Reward-Related Learning in the Human Brain	Paper
	Armita Kaboli	An Efficient Placement and Routing Technique for Fault-tolerant Distributed Embedded Computing	PDF
	greiner
	dale
	cosmin
	masoud
	bjjoyce
	awhite
	erafols

Previous rounds

Round 1: (May 15 - July 20)

Paper suggestions (see also the RLAI paper bank)

Extend this page to add a suggestion, or edit to remove as appropriate. (Click on "Extend this Page" in footer).

eyes on the prize (Nilsson, AI magazine)
reasoning in rats (Tolman and Honzig)
Intelligence without representation, Rodney Brooks
Artificial intelligence meets natural stupidity, by Drew McDermott
i think we should do this animal learning theory paper. Looks very interesting, and it is by Jim Kehoe, who will visit us in the fall.
AdaBoost
Rivest and Schapire (PSR predecessor)
Boyan (FA paper)
Marc Toussaint (Sensorimotor map)

Some of the accepted ICML papers would make great topics for tea-time talks. They are now up on this website:

http://icml2006.org/icml2006/technical/accepted.html Dave, Thu May 25 09:45:13 2006

Who thinks there should be a second round of talks? Anonymous, Mon Jul 17 14:59:42 2006

For this Wednesday, 20 September, I will be doing the Tea Time Talk, entitled: "Is the generation of a learned response linear or catastrophic?" Jim Kehoe, Mon Sep 18 16:47:22 2006

Tea-time Talks