Chargement...
 

Historique: Reservoir Computing

Aperçu de cette version: 28

<h3 id="Participants" class="showhide_heading">Participants</h3>
<div>Nicolas Bredèche, Fei Jiang, Cédric Hartland. Hélène Paugam-Moisy, Julien Perez, Marc Schoenauer, Michèle Sebag<br />
INRIA-Alchemy Hugues Berry, Benoît Siri, Olivier Temam</div>
<div>LIMSI-PS Philippe Tarroux, Jean-Sylvain Liénard, Mathieu Dubois, Sylvain Chevallier</div>
<div>* Ludovic Arnold</div>
<div>* Autres participants extérieurs : Pierre Yger (UNIC, Gif), Yann Chevaleyre (LAMSADE, Paris-Dauphine)</div>
<div>&#160;</div>
<h3>Research Themes</h3>
<div>Reservoir Computing is a generic name that appeared at the end of 2006, gathering new paradigms for computing large recurrent neural networks with random connectivity, such as:</div>
<div>&#160;&#160; - ESN = Echo State Networks (Jaeger, 2001)</div>
<div>&#160;&#160; - LSM = Liquid State Machines (Maass, Natschlager, Markram, 2002)</div>
<div>Since they are dynamical, non-linear systems, recurrent networks are both highly powerful and very hard to tame. Highly powerful for solving complicated engineering tasks: control and modeling complex dynamical systems, temporal pattern recognition, speech recognition, autonomous robotics. Very hard to tame as long as the designer aims to control all the clockwork of the network dynamics for designing efficient learning rules.</div>
<div>The trick of Reservoir Computing is to design an internal network - the so-called "reservoir" - as a large random structure, with sparse connectivity, and to take advantage of its free auto-organization by learning with easy rules a set of "read-out" neurons, in order to extract relevant information from the reservoir dynamics. Neurons involved in such structures can be traditional neurons (threshold, sigmoid units) as well as spiking neurons that take into account the precise timing of spike firing, drawing their inspiration from cognitive processing and memory in the brain.</div>
<div>Understanding and designing efficient models and methods for Reservoir Computing is a matter for research at the border of machine learning (rethinking learning rules with a temporal dimension) and complex systems (optimizing the emergence of a complex dynamical behavior from the structure of the reservoir). Coupling Reservoir Computing and Evolutionary Computation is also a topic of interest for the present working group.</div>
<div>&#160;</div>
<h3>Meetings</h3>
<div>The Reservoir Computing SIG meets every two weeks. See the&#160;<a href="http://tao.lri.fr/tiki-calendar.php">TAO calendar</a> and the&#160;<a href="http://tao.lri.fr/tiki-index.php?page=Reservoir+Computing+-+Private+page">private page</a> (restricted access) for details.</div>
<ul>
<li>16/10/2008 - ESN, LSM or other: which structure and learning rule ? for which task ?</li>
<li>30/10/2008 - Article:&#160; "An experimental unification of reservoir computing methods" read by H.Paugam</li>
<li>12/11/2008 - Invited talk: "Cost-efficiency of human brain networks" - D. Meunier (BMU, Cambridge, UK)</li>
<li>04/12/2008 - Article: " On computational power and the order-chaos phase transition in reservoir computing" read by M. Dubois (LIMSI)</li>
<li>22/01/2009 - <span class="description">Discussion about learning rules used for LSM in literature: p-delta rule for a robot to learn a "gamma" trajectory?</span></li>
<li>28/01/2009 - Simulators in use at TAO: RC toolbox from Ghent (B) team, presented by F. Jiang; Jaeger's toolbox for ESN, presented by J. Perez</li>
<li>12/02/2009 - An introduction to the notion of "Deep Networks" - H. Paugam and M. Sebag. Discussion: which connections with reservoir computing?</li>
<li>05/03/2009 - Article: "<span class="description">A unified architecture for natural language processing: Deep neural networks with multitask learning" read by </span><span class="description">G. Wisniewski and A. Allauzen (LIMSI)</span></li>
<li>20/03/2009 - Article: "<span class="description">A statistical analysis of information processing properties of lamina-specific cortical microcircuit models" read by H. Berry (INRIA-Alchemy)<br />
</span></li>
</ul>
<div>&#160;</div>
<h3>Publications</h3>
<p>F. Jiang, H. Berry and M. Schoenauer. (2008) <a href="http://www.lri.fr/%7Ejiang/papers/PPSN08_SEESN.pdf">Supervised and Evolutionary Learning of Echo State Networks</a>. 10th International Conference on Parallel Problem Solving From Nature, PPSN-2008, September, Dortmund, Germany.</p>
<p>F. Jiang, H. Berry and M. Schoenauer. (2008) Unsupervised Learning of Echo State Networks: Balancing the Double Pole. Genetic and Evolutionary Computation Conference, GECCO-2008, July, Atlanta, GA, USA.</p>
<p>H. Paugam-Moisy, R. Martinez, S. Bengio. (2008) <a href="http://liris.cnrs.fr/Documents/Liris-3399.pdf">Delay learning and polychronization for reservoir computing</a>. Neurocomputing 71(7-9):1143-1158. Elsevier.</p>
<p>F. Jiang, H. Berry and M. Schoenauer. (2007) <a href="http://www.lri.fr/%7Ejiang/papers/ECCS07_EVVON.pdf">Optimizing the topology of complex neural networks</a>. European Conference on Complex Systems (ECCS'07), October, Dresden, Germany.</p>
<p>C. Hartland, N. Bredeche. (2007) Using Echo State Networks for Robot Navigation Behavior Acquisition. 4th IEEE International Conference on Robotics and Biomimetics (ROBIO 2007). pp.201-206. Sanya, China.</p>
<p>A. Devert, N. Bredeche, M. Schoenauer. (2007) Unsupervised Learning of Echo State Networks: A Case Study in Artificial Embryogeny. 8th Int, Conf, on Artificial Evolution (EA'07), Springer LNCS 4926, p278-290, Tours, France.</p>
<p>P. Series and P. Tarroux (1999) Synchrony and delay activity in cortical column models. <cite>Neurocomputing</cite> 26-27:505-510. Elsevier.</p>
<p>&#160;</p>

Historique

Avancé
Information Version
mer. 21 de Oct, 2009 16h51 hpaugam from 129.175.15.11 32
Afficher
mer. 21 de Oct, 2009 16h44 hpaugam from 129.175.15.11 31
Afficher
mer. 21 de Oct, 2009 16h41 hpaugam from 129.175.15.11 30
Afficher
mer. 21 de Oct, 2009 16h08 hpaugam from 129.175.15.11 29
Afficher
lun. 30 de Mar, 2009 14h32 sebag from 129.175.15.11 28
Afficher
ven. 20 de Mar, 2009 12h56 hpaugam from 129.175.15.11 27
Afficher
ven. 20 de Mar, 2009 12h54 hpaugam from 129.175.15.11 26
Afficher
ven. 20 de Mar, 2009 12h51 hpaugam from 129.175.15.11 25
Afficher
ven. 23 de Jan, 2009 15h34 hpaugam from 129.175.15.11 24
Afficher
ven. 23 de Jan, 2009 15h28 hpaugam from 129.175.15.11 23
Afficher