CSCE 478/878 (Fall 2004) Oral Presentation Schedule
Note: At the time of your talk, put your presentation on the web or bring it on
a CD or USB key.
We will have a laptop available, and you should show up early to put your
presentation on the laptop. Doing so before the start of the session as
opposed to right before your talk will speed things up for everyone.
(You may use your own laptop if you wish, but things will go faster if we
use a single machine.)
You are expected to attend as many talks as possible, especially those during
regular class periods. This will affect the subjective portion of your grade.
You should consider your presentation's scheduled time slot final unless you
have a conflict that you absolutely cannot change. This is to prevent
constant modifications of the presentation schedule, which require alterations
to everyone else's schedules and potential problems with room and equipment
reservations.
Each talk will be 20 minutes, plus 5
minutes for questions. Hence a total of 25 minutes is allotted per talk.
However, this is only an approximation! Don't
assume that your talk will start exactly at the time implied by the
given schedule.
- Session 1: Ongoing Research in Machine Learning & New Optimization Methods
Monday, December 6, 2004, 12:30-3:00, Avery 347
- Garrett Michael Matuska
Predicting the Outcome of Horse Races
- Brandon Hauff
Using Concept Lattices for Disjoint Clustering
- Catherine Anderson and Ryan Lim
Collapsing Concavities in ROC Curves via Boosting
- Robert Sprick
An Overview of Active Learning
- Thomas Osugi and Deng Kun
A New Method for Active Learning
- Session 2: Results That Will Make Us All Rich
Tuesday, December 7, 2004, 3:00-5:00, Avery 347
- Jennifer Novotny
Linear Optimzation Methods in Machine Learning
- Kevin Koss
Predicting the Outcome of NFL Games
- Cpt. Jason Beckwith
Reinforcement Learning to Play Five Card Draw
- Chris Bourke
Theoretical Results on learning Disjunctive Normal Form (DNF) Formulas
- Session 3: Advanced Data Representations (Sequence,
Multiple-Instance, Missing Attributes)
Wednesday, December 8, 2004, 12:30-2:30, Avery 347
- Eric Manley
Hidden Markov Models for Information Extraction
- Matt Culver
ANNs for Predicting Time-Series Data
- Tim Perrin
Multiple-Instance Learning
- Tyler Brennfoerder
Handling Noisy and Unspecified Attributes
- Session 4: Performance Enhancement and Genetic Algorithms
Thursday, December 9, 2004, 3:30-5:30, Avery 347
- Ryan Duell and Mark Mager
Genetic Algorithms in Constrained Domains
- Wes McClure
Hardware-Based Genetic Algorithms
- Bret Oltman
Hardware Implementations of Learning Algorithms
- Tim Steiner
Parallelization of Learning Algorithms
-
Contingency Date:
(if Tuesday or Thursday doesn't work): Friday, December 10, 2004, 12:30-2:30,
Avery 256C (note different room)
If the above schedule holds (i.e. if there's no faculty meeting Tuesday
or Thursday) and we do not use this date, then class is cancelled for this
day.
Return to the CSCE 478/878 (Fall 2004) Home Page
Last modified 16 August 2011; please report problems to
sscott.