spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/data.php
Contains all needed files to use the SPiCe data off-line. Including a script to compute the score), that is to say without the iterative submission process used during the competition. If you want to know what was available during the competition, you will find below, for each problem, a file corresponding to the training sample, and 2 files corresponding respectively to the public and the private test sets. More details are given on the participate. Details about the data can be found in this paper.
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/leaderboard.php
To decide the final overall rank of each participant, we summed the scores of the best submission on each problem. Only the scores on the private test samples (a unique submissions) are taken into account for the final results. However, the scores given here are computed using the submission on the public test samples. No limit on the number of submissions). Here is the global ranking of the competition on the public test sets (unlimited number of submissions was allowed). Detailed ranking on each problem.
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/participate.php
This was very important during the competition. Now that it is finished, the data. Are available in a more usual form that allow a off-line computation of the score. The python program submission.py. Is an example of a script for a submission process. At the begining of the script you need to give your user id (the one sent in the registration email) and the problem files. All that is missing then is the learning algorithm and the ranking function! Here is a scenario of a participation:. Is the prefix nu...
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/committee.php
Lancaster University, United Kingdom. Marc G. Bellemare. Google DeepMind, United Kingdom. Xerox Research Center Europe. King's College London, United Kingdom. Aix-Marseille Université, France. Aix-Marseille Université, France. Colin de la Higuera. University of Nantes, France. Universidad Nacional de Córdoba and CONICET, Argentina. Xerox Research Center Europe. INRIA Lille, France. University of Leicester, United Kingdom. Delft University of Technology, The Netherlands. Tilburg University,The Netherlands.
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/index.php
Welcome to the SPiCe webpage! The Sequence PredictIction ChallengE (SPiCe) is an on-line competition about guessing the next element in a sequence of symbols that took place in 2016. Training datasets consist of whole sequences and the aim is to learn a model that allows the ranking of potential next symbols for a given prefix, that is, the most likely options for a single next symbol. You can also have a look at these slides. Please, cite this peer-reviewed paper if you are using the SPiCe data. October...
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/description.php
Details about the competition can be found in this paper. How the success of the learning is evaluated. As the participants submit a ranking of potential 5 next symbols, meaning they give an ordered list of 5 symbols starting from the most probable next symbol, we are using a ranking metric based on normalized discounted cumulative gain. Suppose the test set is made of prefixes y. Suppose now that the distinct next symbols ranking submitted for y. We can compute the following measure for prefix y. We are...
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/baseline.php
We offer different baselines and more importantly scripts to submit result in the iterative way required by the competition. However, now that it is over, an off-line version of the data and the baselines. A python API is available here. A spectral learning baseline in python was developped for the SPiCe competition! Here is the steps to use it:. You first needs to get the Sp2Learning (for SPiCe Spectral Learning) package. The easiest way it by using pip: you just need to wrote. This is the second versio...
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/timeline.php
Base learner(s) and first testing data-sets are available. July 31st 2016, AoE. The competition ends, the results are announced. The short papers from the active participants are due. A special session takes place at ICGI 2016. The winner(s) is (are) expected to submit a paper to the journal special issue that will follow the ICGI 2016 conference.
spice.lif.univ-mrs.fr
SPiCe: Sequence PredIction ChallengE
http://spice.lif.univ-mrs.fr/final_results.php
This page contains the final result of the SPiCe competition. Details about the result can be found in this paper. To decide the final overall rank of each participant, we sumed the scores of the best submission on each problem. The team whose overall score is higher is the winner of the competion. Here is the global ranking of the competition:. Head of the Team. Detailed ranking on each problem. With submission named tttt0801 0449. With submission named ushitora1469525756. Team Markov s Principle. With ...