Publications
August 3rd, 2006 by Francois Rivest
Copyright notice
Abstracts, papers, chapters, and other documents are posted on this site as an efficient way to distribute reprints. The respective authors and publishers of these works retain all of the copyrights to this material. Anyone copying, downloading, bookmarking, or printing any of these materials agrees to comply with all of the copyright terms. Other than having an electronic or printed copy for fair personal use, none of these works may be reposted, reprinted, or redistributed without the explicit permission of the relevant copyright holders.
To obtain a PDF reprint of a particular article, signal your agreement with these copyright terms by clicking on the [pdf] link.
Rivest, F. (2010) Neural Models of Temporal Representations and Reinforcement Learning. Physiology Seminars. Queen’s University, Kingston, On. Invited Presentation
Rivest, F., Kalaska, J.F., & Bengio, Y. (2010, submitted) Conditioning and Time Representation in the Long Short-term Memory Networks.
Rivest, F. (2010) Learning and the partial observability of continuous time. 5th Barbados Workshop on Reinforcement Learning: Sculpting Representation for Reinforcement Learning. McGill University, Barbados. Invited Presentation
Rivest, F. (2009) Modèle informatique du coapprentissage des ganglions de la base et du cortex : L’apprentissage par renforcement et le développement de représentations. Thèse de doctorat, Département d’informatique et de recherche opérationnelle, Université de Montréal. [pdf]
Rivest, F. (2009) Reinforcement and representation learning in the brain. Reasoning and Learning Laboratory, School of Computer Science. McGill University, Montréal, Qc. Invited Presentation
Rivest, F., Kalaska, J.F., & Bengio, Y. (2009) Alternative time representation in dopamine models. Journal of Computational Neuroscience 28(1):107-130. doi: 10.1007/s10827-009-0191-1 [pdf]
Rivest, F. (2009) Computational Neuroscience of Reinforcement Learning. Cognitive Science, Department of Psychology, McGill University, Montreal, Qc. Invited Lecture
Rivest, F., Bengio, Y. & Kalaska, J.F. (2008) Learning timing in reinforcement learning model of dopamine responses in appetitive fixed-delay classical conditioning. Society for Neuroscience Abstracts 378.10. Poster
Rivest, F. (2008) Computational Neuroscience of Reinforcement Learning. Cognitive Science, Department of Psychology, McGill University, Montreal, Qc. Invited Lecture
Rivest, F. (2008) Apprentissage par renforcement. Fondements de l’apprentissage machine, Département d’informatique et de recherche opérationnelle, Université de Montreal, Qc. Invited Lecture
Rivest, F. (2007) Atelier Matlab. Groupe de recherche sur le système nerveux central, Université de Montreal, Qc. Multiple sessions tutorial
Shultz, T.R., Rivest, F., Egri, L., Thivierge, J.-P., & Dandurand, F. (2007) Could Knowledge-based Neural Learning Be Useful in Developmental Robotics? The Case of KBCC. International Journal of Humanoid Robotics (Special Issue on Autonomous Mental Development) 4(2):245-279. [pdf]
Rivest, F. (2007) Computational Neuroscience of Reinforcement Learning. Cognitive Science, Department of Psychology, McGill University, Montreal, Qc. Invited Lecture
Dandurand, F., Shultz, T.R., & Rivest, F. (2007) Complex problem solving with reinforcement learning. In Proceeding of the 6th IEEE International Conference on Development and Learning (ICDL-2007), pp. 157-162. IEEE. [pdf]
Rivest, F. (2007) Real Neurons for Machine Learning. Séminaires UdeM-McGill-MITACS d’Apprentissage Automatique. Université de Montréal, Qc. Presentation
Rivest, F., Kalaska, J.F., & Bengio, Y. (2007) Modèle neuroinformatique du signal dopaminergique et de l’intervalle de temps en conditionnement à délai fixe. L’approche transdisciplinaire des sciences cognitives, ACFAS 2007. Poster
Thivierge, J.-P., Rivest, F., & Monchi, O. (2007) Spiking Neurons, Dopamine, and Plasticity: Timing Is Everything, But Concentration Also Matters. Synapse 61:375-390. [pdf]
Rivest, F. (2006) Modèles computationels des neurones dopaminergiques: Le temps, tant oublié (Partie 2). Séminaires étudiants en sciences neurologiques. Groupe de Recherche sur le Système Nerveux Centrale, Université de Montréal, Orford, QC. Presentation
Rivest, F., Kalaska, J.F., & Bengio, Y. (2006) Model of Time Interval Acquisition in Fixed-Delay Appetitive Classical Conditioning. GRSNC, XVIIIe symposium international, Computational Neuroscience Computationnelle. Poster
Rivest, F. (2006) Neural Basis of Learning. Séminaires UdeM-McGill-MITACS d’Apprentissage Automatique. Université de Montréal, Qc. Presentation
Shultz, T.R., Rivest, F., Egri, L., & Thivierge, J.P. (2006) Knowledge-based learning with KBCC. Proceedings of the Fifth International Conference on Development and Learning ICDL 2006. Department of Psychological and Brain Sciences, Indiana University, Bloomington. [pdf]
Rivest, F. (2005) Modèles computationels des neurones dopaminergiques: Le temps, tant oublié. Retraite du Groupe de Recherche sur le Système Nerveux Centrale. Université de Montréal, Orford, QC. Presentation
Rivest, F., & Shultz, T.R. (2005) Learning with Both Adequate Computational Power and Biological Realism. Proceedings of the 2005 Canadian Artificial Intelligence Conference: Workshop on Correlation Learning, pp. 15-23. University of Victoria, Victoria, BC. [pdf]
Rivest, F. (2005) L’apprentissage par renforcement et le développement d’abstractions dans le cerveau. Examen pré-doctoral, Département d’informatique et de recherche opérationnelle, Université de Montréal. [pdf]
Rivest, F., Bengio, Y, & Kalaska, J.F. (2005) Brain Inspired Reinforcement Learning. In Lawrence K. Saul, Yair Weiss, and Léon Bottou, editors, Advances in Neural Information Processing Systems 17, pp. 1129-1136. MIT Press, Cambridge, MA. [pdf]
Bellemare, M.G., Precup, D., & Rivest, F. (2004) Reinforcement Learning Using Cascade-Correlation Neural Networks. Technical Report RL-3.04. School of Computer Science, McGill University. [pdf]
Rivest, F., & Shultz, T.R. (2004) Compositionality in a Knowledge-based Constructive Learner. Papers from the 2004 AAAI Symposium, Technical Report FS-04-03, pp. 54-58. AAAI Press: Menlo Park, CA. [pdf]
Rivest, F., Bengio, Y., & Kalaska, J.F. (2004) Learning Motor Skills In Unsupervised Sensory Cortex, Reinforced Basal Ganglia, And Semi-Unsupervised Frontal Cortex, Motor Learning & Plasticity Satellite Meeting, Neural Control of Movement Conference. Poster
Rivest, F. (2003) Combiner l’apprentissage non-supervisé à l’apprentissage par renforcement/Combining Unsupervised Learning to Reinforcement Learning. MITACS Quebec Interchange, Mathematics of Information Technology and Complex Systems. Poster
Rivest, F., & Precup, D. (2003). Combining TD-learning with Cascade-correlation Networks. Proceedings of the Twentieth International Conference on Machine Learning, pp. 632-639. AAAI Press. [pdf]
Thivierge, J.-P., Rivest, F., & Shultz, T.R. (2003). A Dual-phase Technique for Pruning Constructive Networks. Proceedings of the IEEE International Joint Conference on Neural Networks 2003, pp. 559-564. [pdf]
Rivest, F. (2003) Apprentissage dans le cerveau. Laboratoire d’Informatique des Systèmes Adaptatifs. Université de Montréal, Qc. Presentation
Shultz, T. R., & Rivest, F. (2003). A tutorial on knowledge-based cascade-correlation. Slideshow
Dandurand, F., Rivest, F., Stolle, M., & Shultz, T. R. (2003). LNSC cascade-correlation simulator applet. Internet applet
Le Dévéhat, Y, Perron, D, Fraysse, O., Dumouchel, P., Landry, R.Jr. & Rivest, F. (2003) Method of Improving Successful Recognition of Genuine Acoustic Authentication Devices. US Patent
Shultz, T. R., & Rivest, F. (2003). Knowledge-based cascade-correlation: Varying the size and shape of relevant prior knowledge. In H. Yanai, A. Okada, K. Shigemasu, Y. Kano, & J. J. Meulman (Eds.), New developments in psychometrics, pp. 631-638. Tokyo: Springer-Verlag. [pdf]
Rivest, F. (2002) Knowledge-Transfer in Neural Network: Knowledge-Based Cascade-Correlation. M.Sc. Thesis, School of Computer Science, McGill University. [pdf]
Rivest, F. & Shultz, T.R. (2002) Application of Knowledge-based Cascade-correlation to Vowel Recognition, IEEE International Joint Conference on Neural Network 2002, pp. 53-58. IEEE Society Press. [pdf]
Shultz, T. R., & Rivest, F. (2001). Knowledge-based cascade-correlation: Size variation of relevant prior knowledge. IMPS-2001: International Meeting of the Psychometric Society (p. 99). Presentation
Shultz, T.R. & Rivest, F. (2001) Knowledge-based Cascade-correlation: Using Knowledge to Speed Learning, Connection Science 13:1-30. [pdf]
Shultz, T.R. & Rivest, F. (2000) Knowledge-based Cascade-correlation: An Algorithm for Using Knowledge to Speed Learning. Proceedings of the Seventeenth International Conference on Machine Learning, pp. 871-878. San Francisco, CA: Morgan Kaufmann. [pdf]
Shultz, T.R. & Rivest, F. (2000) Knowledge-based Cascade-correlation, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Network 2000, pp. V641-V646. Los Alamitos, CA: IEEE Society Press. [pdf]
Rivest, F. (1999) A New Approach to the Use of Prior Knowledge in Neural Networks. Second Annual Undergraduate Summer Workshop in Cognitive Science, Institute for Research in Cognitive Science, University of Pennsylvania. Poster
Computational Neuroscience (and Programming) Blog
Hi Francois,
You seem to be missing the pdf file for the article “Reinforcement Learning Using Cascade-Correlation Neural Networks”, is it possible that you could mail it to me (lukesky [at] diku.dk)?
Best Regards,
Steffen
Thanks, I’ll fix that soon.
Â
Done
HI,
In the paper
“Knowledge-based Cascade-correlation, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Network 2000, pp. V641-V646. Los Alamitos, CA: IEEE Society Press”
You have mentioned that
“Our extension, called knowledge-based cascade-correlation (KBCC) recruits previously learned networks in addition to the untrained hidden units recruited by CC.”
can u please say me what do you mean by the “existing source knowledge” and “previously learned network”. Does it mean the output that we get from the already existing network like Back propogation neural network(BPNN)and are we applying the KBCC to the output of the BPNN? Please help me with the same.
KBCC can reuse prior knowledge when learning a new task instead of starting from scratch. That prior knowledge can be any mathematical function, but in this paper, the sources of knowledge are CC networks trained on similar task before hand. Those networks could have been BPNN, it does not matter to KBCC (see my MSc. Thesis).
As you can see in the Figure 1 of the paper you cite, one such previously trained CC network is recruited inside KBCC architecture. A recruited network in KBCC is process a bit like a hidden layer in the forward pass of a BPNN. That is, all lower layers are process first; then a weighted sum of them is computed, and the resulting vector is passed as input to the recruited network. Finally, the output of the recruited network is used by further KBCC layers as if that recruited network was simply a hidden layer in KBCC.
If you are not familiar with the cascade-correlation architecture, I suggest you to first read Fahlman & Lebiere 1990 (NIPS2). Also, for a more detailed description of KBCC, I suggest you to read our Connection Science 2001 paper (or my MSc. Thesis).
Hi,
Thank you very much for your response. I couldnt cite out the paper you have mentioned “Fahlman & Lebiere 1990 (NIPS2)”. If you have the reference paper, can you please e-mail me the paper.
I did. Good reading!
By the way, all NIPS papers are freely available online at http://books.nips.cc/.