IJCAI'99 Workshop

Frank Hoffmann (fhoffman@cs.berkeley.edu)
Wed, 2 Jun 1999 07:29:03 +0200 (MET DST)

Berkeley Initiative in Soft Computing (BISC)

IJCAI'99 Workshop on


Co-chairs: Lee Giles and Ron Sun

(Stockholm, Sweden, August 1st, 1999)


Sequence learning is an important component of learning in many task
domains: inference, planning, reasoning, robotics, natural language
processing, speech recognition, control, time series prediction, financial
engineering, DNA sequencing, etc. There are many different approaches
towards sequence learning, resulting from different perspectives taken in
different task domains. These approaches deal with somewhat differently
formulated sequential learning problems (for example, some with actions and
some without).

Sequence learning is a difficult task, and more powerful algorithms are
needed in all of these domains. The right approach is to better understand
the state of the art in different disciplines related to this topic first.
Therefore, there seems to be a need to compare, contrast, and combine
different techniques, approaches, and paradigms, to develop more powerful
algorithms. These techniques and algorithms include recurrent neural
networks, hidden Markov models, dynamic programming (reinforcement
learning), graph theoretical models, evolutionary computational models, AI
planning models, rule-based models, etc. We need a gathering that includes
researchers from all of these orientations and disciplines, beyond narrowly
focused topics such as reinforcement learning or neural networks for
sequential processing.

The following questions and issues will be addressed:

1. underlying similarity and difference of different models
1.1 problem formulation (ontological issues)
1.2 mathematical comparisons
1.3 task appropriateness
1.4 performance analysis and bounds

2. new and old models: capabilities and limitations
2.1 theory
2.2 implementation
2.3 performance
2.4 empirical comparisons in various domains

3. hybrid models: approaches, theories and applications
3.1 foundations for synthesis or hybridization
3.2 necessity, advantages, problems, and issues

4. successful sequence learning applications and future extensions
4.1 examples of successful applications
4.2 generalization and transfer of successful applications
4.2 what is needed for enhancing performance


(6 invited and 11 contributed papers)


(Each speaker should leave 5 minutes of their alloted time for questions
and discussions.
Each invited speaker with a 40-minute presentation should include 10 minutes
for discussion.)

9:00-9:05 Opening remarks, Ron Sun and Lee Giles

1. RL and SDM:

9:05-9:45 Manuela Veloso, CMU. "Individual and Team Learning in
Complex Multi-Agent and Adversarial Environments"

9:45-10:25 M. Niranjan, Cambridge Univeristy. "Algorithms for
Sequential Learning tasks"

10:25-10:50 break

10:50-11:10 S. Choi, D. Yeung, N. Zhang. "Hidden-Mode Markov Decision

11:10-11:30 T. Oates, L. Firoiu, P. Cohen. "Clustering Time Series
with Hidden Markov Models and Dynamic Time Warping"

2. Sensory-Motor Sequences:

11:30-11:50 P. Sebastiani, M. Ramoni, and P. Cohen. "Unsupervised
Classification of Sensory Inputs in a Mobil Robot"

11:50-12:10 R. Bapi and K. Doya, "MFM: Multiple Forward Model
Architecture for Sequence Processing"

3. Poster Summaries: (5 minutes each; 12:10-12:30)

N. Rougier, H. Frezza-Buet, F. Alexandre. "Neuronal Mechanisms for
Sequence Learning in Behavioral Modeling"

E. Sang, J. Nerbonne. "Learning Simple Phonotactics"

M. Rosenstein, P. Cohen. "Continuous Categories for a Mobile Robot"

L. Brehelin, O. Gascuel, G. Caraux. "Learning Sequences of Vectors
using Hidden Markov Models with Patterns: Application to Testing
Integrated Circuits"

12:30-2:00 lunch

4. Neural Networks:

2:00-2:40 Alessandro Sperduti, U. of Pisa. "On the Need for a Neural
Abstract Machine"

2:40-3:20 Juergen Schmidhuber, IDSIA. "Continual Prediction through
LSTM with Forget Gates" (Felix Gers, Juergen Schmidhuber, Fred

3:20-3:40 P. Boden, J. Wiles, B. Tonkes. A. Blair. "On the Ability of
Recurrent Nets to Learn Deeply Embedded Structures"

3:40-4:00 break

5. Application-Specific Models:

4:00-4:40 Kevin Lang, NEC Research Institute. "State Merging Algorithms
for DFA Learning"

4:40-5:20 Mohammad Zaki, RPI. "Mining Frequent Sequences "

5:20-5:40 P. Baldi, S. Brunak, P. Frasconi. "Bidirectional Dynamics
for Protein Secondary structure Prediction"

5:40-6:00 A. Nowe, K. Verbeeck. "Distributed Reinforcement Learning,
Load-based Routing"

6. Panel Discussions, and Conclusions:

6:00-6:30 Panel (chaired by Ron Sun and Lee Giles)
(focus on comparing different methods and highlighting interesting

7. Poster Presentations: (posters should be set up before the morning coffee break)

N. Rougier, H. Frezza-Buet, F. Alexandre. "Neuronal Mechanisms for
Sequence Learning in Behavioral Modeling"

E. Sang, J. Nerbonne. "Learning Simple Phonotactics"

M. Rosenstein, P. Cohen. "Continuous Categories for a Mobile Robot"

L. Brehelin, O. Gascuel, G. Caraux. "Learning Sequences of Vectors
using Hidden Markov Models with Patterns: Application to Testing
Integrated Circuits"



To encourage discussions, accepted contributions and discussion topics are
published on the world wide web before the workshop. As a consequence, the
content of all the talks is known beforehand, so that presentations and
discussions can focus on the technical questions.

Hardcopy ``Working Notes" will be available at the workshop (but are
available here online). We are also considering publishing an edited book
after the workshop with a major publisher.

Accessing the workshop papers in postscript:

If you want to participate in the workshop, see:

Committee Members:

Jack Gelfand, Princeton Univeristy
Lee Giles, NEC Research Institute
Marco Gori, U. of Florence
M. Niranjan, Cambridge Univeristy
Ron Sun, U of Alabama/NEC RI
Gerry Tesauro, IBM

Dr. C. Lee Giles (co-chair)
NEC Research Institute, 4 Independence Way, Princeton, NJ 08540, USA
609-951-2642, giles@research.nj.nec.com

Professor Ron Sun (co-chair)
Department of Computer Science, The University of Alabama, Tuscaloosa, AL 35487
609-951-2781, rsun@cs.ua.edu


Frank Hoffmann UC Berkeley
Computer Science Division Department of EECS
Email: fhoffman@cs.berkeley.edu phone: 1-510-642-8282
URL: http://http.cs.berkeley.edu/~fhoffman fax: 1-510-642-5775
If you ever want to remove yourself from this mailing list,
you can send mail to <Majordomo@EECS.Berkeley.EDU> with the following
command in the body of your email message:
unsubscribe bisc-group
or from another account,
unsubscribe bisc-group <your_email_adress>
Do NOT send unsubscribe requests to bisc-group@cs.berkeley.edu

This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
(4) WWW access and other information on Fuzzy Sets and Logic see
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html