This code shows how Hebbian learning mediated by STDP mechanisms could explain the emergence of disparity selectivity in the early visual system. This upload is a snapshot of the code at the time of acceptance of the paper. For a link to a soon-to-come git repository, consult the author's website: www.tusharchauhan.com/research/ . The datasets used in the paper are not provided due to size, but download links and expected directory-structures are. The user can (and is strongly encouraged to) experiment with their own dataset. Let me know if you find something interesting! Finally, I am very keen on a redesign/restructure/adaptation of the code to more applied problems in AI and robotics (or any other field where a spiking non-linear approach makes sense). If you have a serious proposal, don't hesitate to contact me [research AT tusharchauhan DOT com ].
Model Type: Connectionist Network
Cell Type(s): Abstract integrate-and-fire neuron
Model Concept(s): STDP; Synaptic Plasticity; Hebbian plasticity; Vision
Simulation Environment: MATLAB
Implementer(s): Chauhan, Tushar [research at tusharchauhan DOT com]; Montlibert, Alexandre [Alexandre.Montlibert at cnrs.fr]
References:
Chauhan T, Masquelier T, Montlibert A, Cottereau BR. (2018). Emergence of Binocular Disparity Selectivity through Hebbian Learning. The Journal of neuroscience : the official journal of the Society for Neuroscience. 38 [PubMed]