Idemia Identity & Security

Morpho

Idemia Identity & Security France

Safran Identity & Security

Constance Morel

Herve Chabanne

Julien Bringer

Hervé Chabanne

Emmanuel Prouff

Jonathan Milgram

Olivier Clemot

Olivier CLÉMOT

A method for secure learning of parameters of a convolution neural network, CNN, for data classification includes the implementation, by data processing of a first server, including receiving from a second server a base of already classified learning data, the learning data being homomorphically encrypted; learning in the encrypted domain, from the learning database, the parameters of a reference CNN including a non-linear layer (POLYNOMIAL) operating an at least two-degree polynomial function approximating an activation function; a batch normalization layer before each non-linear layer (POLYNOMIAL); and transmitting the learnt parameters to the second server, for decryption and use for classification.

Methods for secure learning of parameters of a convolution neural network, and for secure input data classification

The invention proposes a method comprising the evaluation of a function F obtained by applying to n sub-functions fa first operation, the evaluation comprising:

Multiparty secure calculation method protected against a malevolent party

the invention proposes a method and an associated system for authenticating a user, by means of the redundancy present between several images of a video, the method using garbled circuits, named variant garbled circuits, associated with the alternative bits between the images of the video and a garbled circuit named invariant garbled circuit, associated with the invariant bits between the images of the video, so that the invariant garbled circuit only needs to be evaluated a single time.

Authentication method and system by garbled circuit

A method for calculating a function as a sub-function fof a datum of a client, a sub-function fof a datum, and a product of n indexed sub-functions fof both data by randomly generating, n indexed invertible data rfrom the setwith m being a prime number, generating, for each i from 1 to n, a set for which each element is formed by a product of a datum rwith a possible result of the sub-function of two variables fevaluated in both data, applying an oblivious transfer protocol between the client and a server so that the client recovers, for each i an intermediate datum tequal to: t=r&#xd7;f(x,Y), obtaining, by the client a result T from intermediate data such that: T=f(X')&#xd7;&#x3a0;t, obtaining, by the server a result R from inverted data such that: R=f(Y)&#xd7;&#x3a0;rusing the results in a cryptographic application.

Method for executing a cryptographic calculation and application to the classification by support vector machines

A method for generating a message signature intended to be validated by a verifier server. A client device is configured to hold a private key and a corresponding public key. The method includes offline pre-computation by a hardware security module of a signature token, a result of encryption using a homomorphic encryption function, storage of the signature token generation of the signature of the encrypted message by the homomorphic encryption function from the result of the encryption by the homomorphic encryption function of the private key stored by the client device, of the signature token and of the message. The signature is intended to be validated by the verifier server by the public key.

Method for generating a message signature from a signature token encrypted by means of a homomorphic encryption function

The present invention relates to identification method of an entity executed by an identification system from indexed distance data (d, . . . , d) corresponding to reference entities and comprising:

Growing community of inventors

Issy-les-Moulineaux, France

Constance Morel