Dell Products Inc.

Emc IP Holding Company LLC

Dell & Products L.p.

Paulo Abelha Ferreira

Pablo Nascimento Da Silva

Vinicius Michel Gottin

Adriana Bechara Prado

Tiago Salviano Calmon

Julia Drummond Noce

Maira Beatriz Hernandez Moran

Rômulo Teixeira De Abreu Pinho

Roberto Nery Stelling Neto

Werner Spolidoro Freund

Nalinkumar Mistry

Romulo Teixeira De Abreu Pinho

Adnan Sahin

Ron Arnan

Hugo De Oliveira Barbalho

Iam Palatnik De Sousa

Hagay Dagan

Eric L Caron

Isabella Costa Maia

Karen Braga Enes

David Burth Kurka

Vicente J P Amorim

Victor Da Cruz Ferreira

Vítor Nascimento Lourenço

Claudio Romero

Thais Luca Marques De Almeida

Pedro Fratini Chem

Jonas Furtado Dias

Techniques are disclosed for dynamic edge-weighted quantization. For example, a system can include at least one processing device including a processor coupled to a memory, the at least one processing device being configured to implement the following steps: selecting edge nodes for sampling based on an edge node sampling algorithm configured to use a specified number of edge nodes to be sampled; causing the selected edge nodes to execute a quantization selection procedure; receiving, from the selected edge nodes, identifications of a quantization procedure based on the quantization selection procedure; and selecting a quantization procedure for each edge node, based on the identifications of the quantization procedures for the selected edge nodes.

Edge-weighted quantization for federated learning

One example method includes identifying active device data streams, based on the active device data streams identified, selecting applicable ML models, obtaining a respective classification inference from each of the ML models, wherein each of the classification inferences applies to a respective active device data stream, identifying a current agreement between the respective classification inferences generated by the models, comparing the classification inferences associated with the current agreement with historical classification inferences associated with historical agreements, and based on the comparing, determining whether or not one or more of the data streams comprises a piece of data of an unknown class.

Open set classification based on heterogenous model ensemble in multisensor environments

One example method includes receiving, by a central node, respective data and statistics from each edge node in a group of edge nodes, performing, by the central node, a similarity-based clustering of the edge nodes so that different clusters of edge nodes are defined, sampling, by the central node, edge nodes from each of the clusters to perform a quantization selection method, receiving, by the central node, from the sampled edge nodes, a respective indication of a best-performing quantization method, and electing, by the central node, a quantization method to be used by all the edge nodes.

Similarity-based quantization selection for federated learning with heterogeneous edge devices

An event driven detection model is disclosed. A model operates at a node to identify relevant video data from video streams generated by cameras. Video data that is not relevant is discarded. An objectness score is generated for the relevant video data. The objectness score and position data from position sensors is used to infer an event. When an event is inferred by the model, a decision may be made and performed.

Efficient event-driven object detection at the forklifts at the edge in warehouse environments

One example method includes mapping of a set of environment constraints to various elements of a dataset distillation process, and then performing the dataset distillation process based upon the mapping, and the dataset distillation process is performed in a distributed manner by a group of edge nodes and a central node with which the edge nodes communicate. The environment constraints may include computing resource availability, and privacy constraints concerning the data of the dataset.

Adaptable dataset distillation for heterogeneous environments

One method includes receiving, at a central node, a respective gradient from each node of a cluster, and each of the gradients comprises a respective update to a global machine learning model maintained at the central node, aggregating, by the central node, the gradients to obtain an aggregated gradient for the cluster, and the aggregated gradient is part of a list of aggregated gradients, running, by the central node, a robust aggregation operation on the aggregated gradients in the list to obtain an outlier score for the cluster, and when the outlier score equals or exceeds a specified value, or falls within a specified range of values, identifying the cluster as an outlier, and identifying nodes within the cluster as suspicious.

Lightweight attacker identification method for federated learning with secure byzantine-robust aggregation via clustering

One example method includes performing various operations in a federated learning environment that includes nodes configured and operable to communicate with a central server. The operations include gathering aggregate gradient information from clusters that each include a respective group of some of the nodes, examining the respective gradient information for each of the clusters, and identifying outlier clusters based on the respective aggregate gradient information of those clusters, for each of the clusters identified as an outlier cluster, setting all the nodes in those clusters as suspect nodes, and performing a clustering operation that comprises reassigning each of the suspect nodes to one of the clusters, and a decision to reassign any given one of the suspect nodes is based on an amount of information expected to be obtained as a result of that decision.

Efficient entropy-guided byzantine attacker identification in federated learning

One example method includes identifying features for operational actions that are to be performed at an edge node of an edge computing system. The features include information used by the edge node to execute the operational actions. A first set of the features is received during an input processing stage. The first set of the features are used in the execution of the operational actions. In response to the execution of the operational actions, a second set of the one or more features are obtained, In response to retrieving the first set of the features or in response to obtaining the second set of the features, a local map is updated to include the features. The local map includes a subset of a global map, where the local map and the global map represent a shared representation of a map structure.

Orchestration of action-input representations for decision making in edge environments

One example method includes obtaining input including characteristics of a machine learning (ML) model, specifications of a hardware configuration on which the ML model has been run, and characteristics of a prospective workload, estimating, given the hardware configuration, an energy efficiency of the ML model, using a similarity measure to find peer ML models of the ML model, and each peer ML model is more energy efficient, given the hardware configuration, than the ML model, ranking the peer ML models as suggested alternatives to the ML model, storing, in a database, the characteristics of the ML model, the energy efficiency of the ML model, and the specifications of the hardware configuration, and transmitting the specifications of the hardware configuration, along with workload characteristics, to a recipient.

Recommendation framework of model and hardware for energy-efficient machine learning workloads

Method for model updating in a federated learning environment, including distributing a current model to client nodes; receiving a first set of gradient sign vectors, wherein each gradient sign vector of the first set of gradient sign vectors is received from one client node; generating a first updated model based on the first set of gradient sign vectors; distributing the first updated model to the plurality of client nodes; storing a first shape parameter and a second shape parameter; receiving, in response to distributing the first updated model, a second set of gradient sign vectors, wherein each gradient sign vector of the second set of gradient sign vectors is received from one client node; generating a second updated model based on the second set of gradient sign vectors, the first shape parameter, and the second shape parameter; and distributing the second updated model to the plurality of client nodes.

Fast converging gradient compressor for federated learning

Peer-based positioning of nodes in an environment is disclosed. A node is able to determine its own position using sources. The sources may include fixed and/or mobile beacons. Each of the sources is associated with a distance error and a position error. These errors, along with distance measurements between the node and each of the sources, are used to determine a position and a total positioning error, which reflects the accuracy of the determined position.

Peer-based positioning of mobile devices

One example method includes receiving at a client node of a federation a global machine-learning model that is to be trained by the client node using a training dataset that is local to the client node. In response to receiving the global machine-learning model, determining at the client node if the global machine-learning model is trending toward an overfitted state using a validation dataset. The overfitted state indicates that the global machine-learning model has not been received from a server that is part of the federation because of a client isolation attack. In response to determining that the global machine-learning model is trending towards the overfitting state, causing the client node to leave the federation. In response to determining that the global machine-learning model is not trending towards the overfitted state, training the global machine-learning model using the training dataset to thereby update the global machine-learning model.

Detecting client isolation attacks in federated learning through overfitting monitoring

One example method includes enabling a user to define a query that comprises a hybrid search syntax (HSS), and the HSS is in the form of a single string that comprises a lexical component and a semantic component, receiving the query from the user, parsing the lexical component and the semantic component into respective groups of one or more sub-components, performing a lexical search based on the sub-components of the lexical component, and performing a semantic search based on the sub-components of the semantic component, ranking, separately from each other, results of the lexical search and results of the semantic search, combining the results of the lexical search and the results of the semantic search to create aggregate results, and re-ranking the aggregate results.

Advanced hybrid search syntax for power user content retrieval

Systems and methods for decoupling information into multiple levels of detail based on semantics while providing a compact global representation structure. A tree-based graph represents objects in an environment. The top node provides complete environment information at the lowest level of detail. Leaves of the tree-based graph complement the parent nodes with more detailed local information. This achieved efficient processing, communication, and data compaction.

So-map: a semantic-aware algorithm for optimizing the representation structure of Octomaps

Techniques for edge-enabled trajectory map generation are disclosed herein. For example, a method can include segmenting one or more node trajectories based on data collected at a node, determining one or more initial trajectories based on the segmented node trajectories, and generating a map including trajectories generated by associating attribute data and event data with the initial trajectories.

Growing community of inventors

Rio de Janeiro, Brazil

Paulo Abelha Ferreira