Abstract: A method for monitoring a person of interest in video that may be generated by multiple cameras in different locations. A sequence of images along with data such as time and location, called a tracklet, may be generated for the person of interest. This reference tracklet may then be compared to other tracklets received at later times to determine if the person of interest appears in these other tracklets. Analyses may include locating persons in the video images, calculating person signatures, and comparing signatures to signatures of the person of interest. Analyses may also include filters for possible locations of the person of interest based on how quickly this person could move from one location to another, and filters based on color of the person of interest's clothing. Possible matches may be confirmed by an operator, while matches with high confidence may be automatically identified as the person of interest.

Abstract: A method for automatically providing an electronic training plan to each individual of a targeted group of individuals called “irregulars”, the targeted group including individuals who do not perform at least one activity on a periodic basis, the targeted group being a part of a larger group called set of individuals, the set of individuals including another group called “regulars”, disjointed from the group of irregulars and including individuals who do perform at least one activity on a periodic basis, each individual of the set of individuals being characterized by a set of physiological data and a set of medical data.

Abstract: An iterative method for monitoring a computing device characterized by metric data to be monitored, including, for each iteration, of collecting metric data over a predetermined interval of time, detecting a seasonality pattern of said metric data over said predetermined interval of time, determining an interval-specific model representing the detected seasonality pattern, calculating modelled data using said determined model and the collected metric data, comparing the calculated modelled data with the collected metric data to calculate a score characterizing the difference between the calculated modelled data and the collected metric data, calculating an anomaly likelihood for each data of the collected metric data using the calculated score, detecting an anomaly on a data when probability that the value of said data is an anomaly is greater than a predetermined threshold.

Abstract: A quality control method of a hardware and software computing infrastructure by a quality control computing system. The quality control method including connecting the hardware and software computing infrastructure to a blockchain database by a management and automation software, mining by the computing resources of a blockchain from the blockchain database using the management and automation software, and monitoring performances of the mining operations and recording values of performances in a database of the quality control computing system. The quality control computing system for the hardware and software computing infrastructure is also disclosed to implement the quality control method.

Abstract: This invention relates to a method of using medical data related to patients suffering a given disease, comprising: a phase of gathering (6) medical data related to patients suffering said given disease, from different hospital databases (65 to 68) storing under different data formats said medical data within non-anonymized files, a phase of storing, in a big data database (1), said gathered medical data, anonymized and in a single standardized format, a phase of using (7, 8) said big data database (1): by performing big data processing functions (21, 22) as well as by performing simple raw data extraction (23) from said big data database and direct visualization (24) of said extracted raw data, to generate one or more global rules governing categories of patients suffering said given disease as well as to improve personal medical care for a specific patient suffering said given disease.

Abstract: A method of managing a charge of a plurality of vehicles on a site, the site including a plurality of charging points and a gateway, the gateway being connected between an electricity grid of an electricity provider, configured to provide electricity to the gateway, and the plurality of charging points, the gateway including a controller and an electrical storage device that stores electrical energy. The method, operated by the controller, includes determining the electricity need of the plurality of charging points at a given time, if the electricity need of the plurality of charging points is greater than zero, determining the electrical consumption of the site, and if the electrical consumption of the site is greater than a predetermined site limit, providing electrical energy to the plurality of charging points using electricity at least partially received from the electrical storage device.

Abstract: The invention relates to a method for quantizing a deep neural network including several layers, previously trained during a training phase determining for each layer a set of weights. The method includes a phase of quantizing the deep neural network including determining a disruption limit value of at least one weight of the weight set of the layer, beyond which the output of the deep neural network is erroneous, determining, for a target inference precision of the neural network, and from the disruption limit value, an adjustment limit value of at least one weight of the set of weights, and decreasing an arithmetic precision of at least one weight of the set of weights as a function of the adjustment limit value. The invention also relates to a computer program, a device implementing such a method, and a deep neural network obtained by such a method.

Abstract: The invention relates to a method (100) for tracking a target object in an image stream captured by a camera, at a capture frequency, Fc, said method (100) comprising several iterations of a tracking phase (110), implemented individually for several images, referred to as processed, of said image stream, and comprising the following steps: detecting (114) at least one object, and its position, in the processed image, and identifying (116-124) said target object among the at least one object detected in said processed image; characterized in that said tracking phase (110) is carried out at a detection frequency, Fs, lower than said capture frequency Fc, so that two images processed during two successive iterations of the tracking phase are separated by at least one non-processed image to which said tracking phase is not applied. It also relates to a computer program, a device, and a system implementing such a method.

Abstract: A method for simulating the combustion of a fluid in a combustion chamber, for the design of said combustion chamber, which includes: discretizing the space of the chamber into a given mesh; training a neural network by means of a learning set associating a graph corresponding to the mesh, the vertices of which have, as a value, progress variables predicted by a computational fluid dynamics simulation with local combustion quantities at these vertices; and an iterative simulation phase, where: the values predicted by the neural network of a local combustion quantity at the vertices of the mesh are provided as input to a solver, in order to obtain a value of a progress variable at each vertex of said mesh, and a graph corresponding to the vertices of the mesh is provided to said neural network, each vertex having a corresponding value of said progress variable, obtained by said solver, in order to obtain predicted values of the local combustion quantity at said vertices.

Abstract: The invention relates to a computer implemented method for designing a firmware including writing the firmware version, and providing the firmware version with a version data identifying the firmware version. The method also includes providing the firmware version with a second data, called rollback data, the second data provided to be compared to the version data of other versions of the firmware, and authorizing, and indicating a limit for, older firmware versions to which the firmware may be downgraded. The invention further relates to a firmware obtained by such a method and a method for modifying the version of a firmware installed on a platform and designed by such a method.

Abstract: A computing unit configured to be mounted in a HPC cabinet comprising an external body and an internal body that comprises secondary electronic components. The internal body being releasably coupled with the external body in order to allow extraction of the internal body when the external body is secured fixedly to the HPC cabinet. The computing unit also includes at least one front lever pivotally coupled to the internal body and configured to latch with a latching wall of the external body and to latch with the bottom wall of the external body in order to releasably couple the internal body with the external body.

Abstract: An inter-process communication method for remote processes. The method implements a resource allocation mechanism via a token system that allows a communication operation to be performed only when the target process of the operation has available resources. Each operation costs one token. Tokens are consumed locally and obtained by remote request. This maximizes chances of successful operations and minimizes the bandwidth used to manage these operations.

Abstract: A network security system that analyzes data from network attacks to determine which attacks came from the same attacker, even if the attacker tries to disguise its identity by spreading attacks out over time and attacking from multiple IP addresses. Intrusion detection systems or firewalls may log data for each attack, such as the time of the attack, the type of attack, and the source and target addresses. Embodiments may augment this data with derived attributes that may profile the attacker's behavior. For example, some attackers may spread out attacks over time, but always attack on the same day of the week; some attackers may spread out attacks over different IP addresses, but these addresses may all be in the same country. The original and augmented data may be clustered using an algorithm such as DBSCAN, and each attacker may be identified with one of the resulting clusters.

Abstract: This invention relates to a method of building a hybrid quantum-classical computing network, comprising: a first step of transformation of an application composed of services into a Petri net including both Petri places (8, 9) and Petri transitions (81, 82, 91-94) between said Petri places (8, 9), any said Petri place (8, 9) corresponding to: either a first type building block corresponding to any quantum processing unit (8) which processes a job into a result, or a second type building block corresponding to any plugin unit (9), which converts a job into another job and/or a result into another result, any Petri transition (81, 82, 91-94) corresponding to any link between two building blocks (8, 9), all said links (81, 82, 91-94) being formatted so as to make any building block (8, 9) interchangeable, a second step of transformation of said Petri net into a hybrid quantum-classical computing network, replacing any building block by its corresponding unit (8, 9), interconnecting all said corresponding units (

Abstract: Method for implementing a graph (G) comprising a plurality of vertices (V) and links (E) between the vertices, a set (R) being a collection of subsets (Ri) of said a given number of vertices (Rik) comprising: in said set (R), selecting subsets (Ri, Rj), called pre-selected subsets, such that a tree (Ti, Tj) is associated respectively to said tree (Ti, Tj), said associated trees (Ti, Tj) being pairwise disjoint; comparing the number of vertices (Rik) associated to each of the pre-selected subset, among the pre-selected subsets, choosing the subset for which the number of vertices is the highest

Abstract: Method for implementing a graph (G) comprising a plurality of vertices (V) and links (E) between the vertices, a set (R) being a collection of subsets (Ri) of said a given number of vertices (Rik) comprising: in said set (R), selecting subsets (Ri, Rj), called pre-selected subsets, such that a tree (Ti, Tj) is associated respectively to said subset (Ri, Rj), said associated trees (Ti, Tj) being pairwise disjoint and constructing (301) each tree (Ti); associating a weight to each said tree (Ti); choosing the subset for which the tree (Ti) has the highest weight.

Abstract: Method for implementing a quantum circuit comprising a plurality of qubits as well as operators executed on said qubits, said operators comprising a sequence of ? 4 Pauli rotation gates, a surface code layout comprising an arrangement of said quantum circuit on a quantum chip, the arrangement comprising at least a tree with a plurality of subtrees, at each rotation gate corresponding a subtree of the tree, the method comprising: generating iteratively a directed acyclic graph of said quantum circuit, a front layer of the DAG being a set of rotations that can be effectively implemented at each iteration, and selecting in said front layer of the DAG a subset, called selected subset, of said set of rotations among subsets, called non intersecting subsets, in which the subtrees are arranged not to intersect.

Abstract: The invention relates to a method (200) for training a dynamic-architecture convolutional neural network for semantic image segmentation, comprising a phase (102) of auto-annotation of images, to constitute a database (B1), called the first database, of auto-annotated images, comprising the following steps: applying (104) at least one transformation to at least one source image, and adding (106), to each transformed image thus obtained, a tag, called the first tag, corresponding to said transformation; said method (200) comprising at least one iteration of a phase (110) of training said neural network with the first image database (B1), and a second image database (B2) annotated with a tag, called a second tag, relating to the content of said image. It also relates to a computer program and a device implementing such a method, and a deep-learning dynamic-architecture convolutional neural network obtained by such a method.

Abstract: The invention relates to a data processing method for preparing a dataset that includes a processor that receives a first plurality of data input streams to prepare an output dataset. The plurality of data input streams and the output dataset are different. The method includes standardizing the plurality of data input streams, encoding the normalized data, preprocessing missing data, and transmitting a preprocessed dataset. The invention further relates to a data processing system, and a recording medium on which the data processing program is recorded.

Abstract: An aspect of the invention relates to a method for a plurality of clients to access a file in a distributed file system, the file being replicated on at least one other server, the method comprising the steps of: Receiving, by an access management device, a request to access the part of the file sent by a first client from the plurality of clients, Selecting a first server based on at least one parameter, Authorization the first client to access the file stored by the first server Receiving, following authorization, a request to access the same file sent by a second client, the second client being different from the first client, Selecting a second server based on the parameter, the second server being different from the first server, Authorizing the second client to access the file stored by the second server.