Workshop on Machine Learning Open Source Software 2006
Important Dates
- Submission Date: October 28th, 2006
- Notification of Acceptance: November 8th, 2006 - passed
- Workshop date: December 9th, 2006
Description
Open source tools have recently reached a level of maturity which makes them suitable for building large-scale real-world systems. At the same time, the field of machine learning has developed a large body of powerful learning algorithms for a wide range of applications. Inspired by similar conferences in bioinformatics (BOSC) or statistics (useR), our aim is to build a forum for open source software in machine learning.
The workshop's goal is to foster a vibrant community that provides a viable open source machine learning environment. Ultimately, open source machine learning software should be able to compete with existing commercial closed source solutions. Towards this end, it is not enough for us to just bring existing and freshly developed toolboxes and algorithmic implementations to people's attention, but to initiate further collaboration towards moving from isolated solutions to a collection of tools which are designed to work with one another. Far from requiring integration into a single package, we believe that this kind of interoperability can also be achieved in a collaborative manner, which is especially suited to open source software development practices.
The contributed talks are accompanied by a mandatory demonstration of the software, which in turn must adhere to an approved Open Source License.
Program:
The workshop will be a mixture of talks (including a mandatory demo of the software) and panel/open/hands-on discussions.
Morning session: 7:30am - 10:30am
7:30am Introduction, Organizers
7:40am Scientific Python, Travis Oliphant (PASCAL invited speaker)
8:20am The PLearn machine learning library, Pascal Vincent and Nicolas Chapados
8:40am SVM-lin - Fast Linear SVM Solvers for Supervised and Semi-supervised Learning, Vikas Sindhwani
8:50am Blaise: A Toolkit for High-Performance Probabilistic Interference, Keith Bonawitz and Vikash Mansinghka
9:00am coffee break
9:15am MALLET and GRMM: Discriminative training toolkits for natural language, Charles Sutton and Andrew McCallum
9:25am OpenDP a free Reinforcement Learning toolbox for discrete time control problems, Sylvain Gelly
9:35am ELEFANT: Efficient Learning, Large-scale Inference, and Optimization Toolkit, Kishor Gawande
9:45am LHOTSE Toolbox for Adaptive Statistical Models, Matthias Seeger
9:55am Poster Session and Demonstrations till 10:30am or beyond
Afternoon session: 3:30pm - 6:30pm
3:30pm Weka: Data Mining Software in Java, Geoff Holmes (PASCAL invited speaker)
4:00pm Bayes Blocks: A Python Toolbox for Variational Bayesian Learning, Antti Honkela
4:15pm CVXOPT: A Python Package for Convex Optimization, Lieven Vandenberghe (PASCAL invited speaker)
4:45pm coffee break
5:00pm Lush - Lisp Universal SHell, Yann LeCun and Leon Bottou
5:20pm Free Software Licenses Considered Harmful, Fernando Pereira
5:30pm Panel discussion, How do we foster a Machine Learning Open Source Community? (e.g. Open Source licenses, reproducibility of experiments, and integrating tools from different projects.)
6:20pm Closing remarks, Organizers
Program Commitee
- Jason Weston (NEC Princeton, USA)
- Gökhan BakIr (Google Research, Switzerland)
- Alexander Smola (NICTA Canberra, Australia)
- Gunnar Rätsch (FML Tuebingen, Germany)
- Lieven Vandenberghe (University of California LA, USA)
- Chih-Jen Lin ( National Taiwan University Taipei, Taiwan)
- William Stafford Noble (Department of Genome Sciences Seattle, USA)
- Klaus-Robert Mueller (Fraunhofer Institute First, Germany)
Call for Contributions (closed)
The organizing committee is currently seeking abstracts for talks at MLOSS 2006. MLOSS is a great opportunity for you to tell the community about your use, development, or philosophy of open source software in machine learning. This includes (but is not limited to) numeric packages (as e.g. R,octave,numpy), machine learning toolboxes and implementations of ML-algorithms. The committee will select several submitted abstracts for 20-minute talks. Accepted abstracts will be published on this web site.
If you are interested in speaking at MLOSS, please send us an email at mloss@tuebingen.mpg.de before October 28th, 2006:
- an abstract (1 page A4)
- a URL for the project page, if applicable
- information about the open source license used for your software or your release plans.
Note: Submissions must adhere to a recognized Open Source License (cf. http://www.opensource.org/licenses/ ) and the source code must be available at the time of presentation.
Organizers:
- Soeren Sonnenburg, Mikio Braun
Fraunhofer FIRST Kekulestr. 7, 12489 Berlin, Germany
- Cheng Soon Ong
MPI for Biological Cybernetics and Friedrich Miescher Laboratory of the Max Planck Society Spemannstrasse 39, 72076 Tuebingen, Germany
- S.V.N. Vishwanathan
National ICT Australia, Canberra, Australia.