Abstract: This method for training a binarized neural network, also called BNN, including neurons, with a binary weight for each connection between two neurons, is implemented by an electronic circuit and comprises: a forward pass including calculating an output vector by applying the BNN on an input vector; a backward pass including computing an error vector from the calculated output vector, and calculating a new value of the input vector by applying the BNN on the error vector; a weight update including computing a product by multiplying an element of the error vector with one of the new value of the input vector, modifying a latent variable depending on the product; and updating the weight with the latent variable; each weight being encoded using a primary memory component; each latent variable being encoded using a secondary memory component having a characteristic subject to a time drift.

Abstract: A device for generating a random electric signal, including an input duct, an output duct, a generator of magnetic particles generating magnetic particles in the input duct, a diffusion chamber connected to the input duct and the output duct, wherein the diffusion chamber is designed to diffuse the generated magnetic particles, a displacement unit for displacement of the generated magnetic particles towards the diffusion chamber, and a converter that is designed to generate an electrical signal proportional to a characteristic, wherein the characteristic is the particle density in the diffusion chamber or the passage of magnetic particles at a predetermined location of an output duct connected to the diffusion chamber.

Abstract: Disclosed is a modular stochastic machine capable of carrying out probability calculations and including: at least one stochastic distribution module corresponding to a plurality of random variables and capable of receiving an input of values of specified variables in order to return, as output, a representation of the distribution of at least one non-specified variable determined by the values of the specified variables; and at least two stochastic variable modules each corresponding to a single random variable and including: both a stochastic multiplier capable of receiving, as input, representations of distributions in order to return, as output, a representation of the product distribution, as well as a stochastic proportional normalizer capable of receiving, as input, a representation of a distribution in order to return, as output, a proportionally normalized representation.

Abstract: A device for generating a random electric signal, including an input duct, an output duct, a generator of magnetic particles generating magnetic particles in the input duct, a diffusion chamber connected to the input duct and the output duct, wherein the diffusion chamber is designed to diffuse the generated magnetic particles, a displacement unit for displacement of the generated magnetic particles towards the diffusion chamber, and a converter that is designed to generate an electrical signal proportional to a characteristic, wherein the characteristic is the particle density in the diffusion chamber or the passage of magnetic particles at a predetermined location of an output duct connected to the diffusion chamber.

Abstract: A neuromorphic system comprises a set of at least one input neuron, a set of at least one output neuron and a synaptic network formed from a set of at least one variable-resistance memristive component, said synaptic network connecting at least one input neuron to at least one output neuron, the resistance of the at least one memristive component being adjusted by delivering to the synaptic network write pulses generated by the at least one input neuron, and return pulses generated by the at least one output neuron, the characteristics of the write and return pulses being deduced from the intrinsic characteristics of the at least one memristive component so that the combination of a write pulse and a return pulse in the at least one memristive component results in a modification of its resistance according to a learning rule chosen beforehand.

Abstract: An unsupervised learning method is provided implemented in an artificial neural network based on memristive devices. It consists notably in producing an increase in the conductance of a synapse when there is temporal overlap between a pre-synaptic pulse and a post-synaptic pulse and in decreasing its conductance on receipt of a post-synaptic pulse when there is no temporal overlap with a pre-synaptic pulse.

Abstract: A neuromorphic system comprises a set of at least one input neuron, a set of at least one output neuron and a synaptic network formed from a set of at least one variable-resistance memristive component, said synaptic network connecting at least one input neuron to at least one output neuron, the resistance of the at least one memristive component being adjusted by delivering to the synaptic network write pulses generated by the at least one input neuron, and return pulses generated by the at least one output neuron, the characteristics of the write and return pulses being deduced from the intrinsic characteristics of the at least one memristive component so that the combination of a write pulse and a return pulse in the at least one memristive component results in a modification of its resistance according to a learning rule chosen beforehand.

Abstract: An unsupervised learning method is provided implemented in an artificial neural network based on memristive devices. It consists notably in producing an increase in the conductance of a synapse when there is temporal overlap between a pre-synaptic pulse and a post-synaptic pulse and in decreasing its conductance on receipt of a post-synaptic pulse when there is no temporal overlap with a pre-synaptic pulse.