NEUROBOT

The project aimed to develop new, adaptive and robust methods for the control of intelligent autonomous robots, based on spiking neural networks. The project's most important results were several new supervised learning rules for spiking neural networks, a new class of spike train metrics, a method for estimating oscillation strength in neural networks, and a mechanism for frequency control in neuronal competition models.

The project was funded by the Romanian Goverment (National Authority for Scientific Research) through PNCDI II / Partnerships Programme, for the period September 2007 - December 2010. Due to the financial crisis, the project's funding was temporarilly suspended for about one year during 2009.

The project was led by Coneural and had as partners Babes-Bolyai University of Cluj-Napoca and Transilvania University of Brasov.

Project team

Active members of the project team:

 

Results

Software

Robby: A robotic framework

The oscillation score

The modulus-metric for spike trains

 

Papers in international scientific journals

C. V. Rusu and R. V. Florian (2014), A new class of metrics for spike trains. Neural Computation, 26(2), 306–348.

R. V. Florian (2012), The chronotron: a neuron that learns to fire temporally-precise spike patterns. PLoS ONE 7 (8), e40233.

R.C. Muresan, O.F. Jurjut, V.V. Moca, W. Singer, D. Nikolic (2008), The oscillation score: An efficient method for estimating oscillation strength in neuronal activity. Journal of Neurophysiology 99, pp. 1333-1353.

D. Nikolic, V.V. Moca, W. Singer and R.C. Muresan (2008), Properties of multivariate data investigated by fractal dimensionality. Journal of Neuroscience Methods 172 (1), pp. 27-33.

R. V. Florian (2010), Challenges for interactivist-constructivist robotics, New Ideas in Psychology, 28 (3), pp. 350-353.

R. Curtu, A. Shpiro, N. Rubin, J. Rinzel (2008), Mechanism for frequency control in neuronal competition models. SIAM Journal on Applied Dynamical Systems 7 (2), pp. 609-649.

 

International conferences - full papers

R. V. Florian (2008), Tempotron-like learning with ReSuMe. In V. Kurkova et al. (eds.), Proceedings of the 18th International Conference on Artificial Neural Networks (ICANN 2008), Prague, Czech Republic. Lecture Notes in Computer Science 5164, pp. 368-375, Springer, Berlin / Heidelberg.

V.V. Moca, D. Nikolic, R.C. Muresan (2008), Real and modeled spike trains: Where do they meet? In V. Kurkova et al. (eds.), Proceedings of the 18th International Conference on Artificial Neural Networks (ICANN 2008), Prague, Czech Republic. Lecture Notes in Computer Science 5164, Springer, Berlin / Heidelberg.

H. Jakab, L. Csató (2009), Q-Learning and policy gradient methods. Proceedings of the International Conference on Knowledge Engineering, Principles and Techniques, KEPT 2009, Cluj-Napoca (Romania), July 2-4, 2009, pp. 175-178. Studia Universitatis Babes-Bolyai, Informatica, special issue KEPT.

L. Csató, B. A. Bócsi (2009), Dirichlet process–based component detection in state-space models, ESANN'2009 proceedings, European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning. Bruges (Belgium), 22-24 April 2009; pp. 491-496.

B. A. Bócsi, L. Csató (2008), Visual tracking with filtering algorithms. IEEE International Conference on Intelligent Computer Communication and Processing 2008, Cluj-Napoca, România, pp. 269-274.

L. Csató, Z. Bodó (2009), Decomposition Methods for Label Propagation. Proceedings of the International Conference on Knowledge Engineering, Principles and Techniques, KEPT 2009, Cluj-Napoca (Romania), July 2-4, 2009, pp. 127-130. Studia Universitatis Babes-Bolyai, Informatica, special issue KEPT.

B. Reiz, L. Csató (2009), Bayesian Network Classifier for Medical Data Analysis. International Journal of Computers, Communications & Control, 4 (1), pp. 65-72.

B. Reiz, L. Csató, D. Dumitrescu (2008), Prüfer number encoding for genetic Bayesian network structure learning algorithm, International Symposium on Symbolic and Numeric Algorithms for Scientific Computing, Timisoara, România, IEEE Computer Society Press, pp. 83-86.

L. Sasu (2009), FAMR: A Neural Network with Relevance Factor for Data Mining, Proceedings of the 1st International Conference on Manufacturing Engineering, Quality and Production Systems (MEQAPS '09), pp. 135-140.

 

International conferences - posters, abstracts

C. V. Rusu and R. V. Florian (2010), A new spike train metric. BMC Neuroscience, 11 (Suppl 1), p. 169.

C. V. Rusu and R. V. Florian (2009), Exploring the link between temporal difference learning and spike-timing-dependent plasticity. Eighteenth Annual Computational Neuroscience Meeting: CNS*2009. BMC Neuroscience 2009, 10 (suppl. 1), p. 201.

R. V. Florian and C. V. Rusu (2009), Temporal difference learning does not always lead to STDP. Frontiers in Systems Neuroscience. Conference Abstract: Computational and systems neuroscience

R. V. Florian (2008), Relating reinforcement learning and STDP. Cosyne workshop on Spiking Networks and Reinforcement Learning, Snowbird, Utah, SUA.

R. V. Florian (2007), Neuromodulation of STDP and reinforcement learning, in K. Josic, M. Matias and J. Rubin (eds.), International Workshop on Coherent Behavior in Neuronal Networks, Mallorca, Spain, p. 33.

R. Andonie, A. Cataron, L. M. Sasu (2008), Fuzzy ARTMAP with feature weighting. In A. Gammerman (ed.), Proceedings of Artificial Intelligence and Applications Conference (AIA 2008), Innsbruck, Austria. Acta Press, Calgary, Canada.

 

Movies

Robby in action: