Alerta

REFLEXIVE MODEL GRADIENT-BASED RULE LEARNING

Absstract of: US2025225386A1

Systems, devices, methods, and computer-readable media for reflexive model generation and inference. A method includes receiving probabilistic rules of a reflexive model that correlate evidence with existence of an event of interest, training, based on ground truth examples of evidence and respective labels indicating whether the event of interest is/was/will be present or not, a neural network (NN) to encode the probabilistic rules and learn respective probabilities for the probabilistic rules, and providing, by the NN and responsive to new evidence, an output indicating a likelihood the event of interest exists.

INTEGRATION OF LEARNED DIFFERENTIABLE LOSS FUNCTIONS IN DEEP LEARNING MODELS

Absstract of: US2025225384A1

Systems and methods are disclosed herein for training a model with a learned loss function. In an example system, a first trained neural network is generated based on application of a first loss function, such as a predefined loss function. A set of values is extracted from one or more of the layers of the neural network model, such as the weights of one of the layers. A separate machine learning model is trained using the set of values and a set of labels (e.g., ground truth annotations for a set of data). The machine learning model outputs a symbolic equation based on the training. The symbolic equation is applied to the first trained neural network to generate a second trained neural network. In this manner, a learned loss function can be generated and used to train a neural network, resulting in improved performance of the neural network.

REMOTE DISTRIBUTION OF NEURAL NETWORKS

Absstract of: US2025225614A1

Remote distribution of multiple neural network models to various client devices over a network can be implemented by identifying a native neural network and remotely converting the native neural network to a target neural network based on a given client device operating environment. The native neural network can be configured for execution using efficient parameters, and the target neural network can use less efficient but more precise parameters.

Generation Of Graph-Based Dense Representations Of Events Of A Nodal Graph Through Deployment Of A Neural Network

Absstract of: US2025225183A1

A computerized method is disclosed that includes operations of receiving a plurality of alerts, generating a graph-based dense representation of each alert of the plurality of alerts including processing of each alert with a neural network, wherein a result of processing an individual alert by the neural network is a graph-based dense representation of the individual alert, computing relatedness scores between at least a subset of the plurality of alerts, and generating a graphical user interface illustrating a listing of at least a subset of the plurality of alerts, wherein the graphical user interface is configured to receive user input corresponding to selection of a first alert, wherein the graphical user interface is rendered on a display screen. Additionally, an additional operation may include training the neural network to produce graph-based dense representations, wherein the training is performed on a corpus of metapaths.

INTEGRATION OF LEARNED DIFFERENTIABLE LOSS FUNCTIONS IN DEEP LEARNING MODELS

Absstract of: EP4583004A1

Systems and methods are disclosed herein for training a model with a learned loss function. In an example system, a first trained neural network is generated based on application of a first loss function, such as a predefined loss function. A set of values is extracted from one or more of the layers of the neural network model, such as the weights of one of the layers. A separate machine learning model is trained using the set of values and a set of labels (e.g., ground truth annotations for a set of data). The machine learning model outputs a symbolic equation based on the training. The symbolic equation is applied to the first trained neural network to generate a second trained neural network. In this manner, a learned loss function can be generated and used to train a neural network, resulting in improved performance of the neural network.

METHOD AND SYSTEM FOR AUGMENTED SPEECH EMBEDDINGS BASED AUTOMATIC SPEECH RECOGNITION

Absstract of: EP4583099A1

Though several data augmentation techniques have been explored in the signal or feature space, very few studies have explored augmentation in the embedding space for Automatic Speech Recognition (ASR). The outputs of the hidden layers of a neural network can be seen as different representations or projections of the features. The augmentations performed on the features may not necessarily translate into augmentation of the different projections of the features as obtained from the output of the hidden layers. To overcome the challenges of the conventional approaches, embodiments herein provide a method and system for augmented speech embeddings based automatic speech recognition. The present disclosure provides an augmentation scheme which works on the speech embeddings. The augmentation works by replacing a set of randomly selected embeddings by noise during training. It does not require additional data, works online during training, and adds very little to the overall computational cost.

CPU DEEP NEURAL NETWORK INFERENCE METHOD BASED ON IDLE CPU RESOURCES TO MAXIMIZE RESOURCE UTILIZATION IN LEARNING CLUSTERS

Nº publicación: KR20250102720A 07/07/2025

Applicant:

성균관대학교산학협력단

Absstract of: WO2025143765A1

A deep neural network inference method based on an idle CPU resource according to the present specification may comprise the steps of: when a training task is reserved and all GPUs are used, classifying a CPU core into a training task-specific group, and classifying an unallocated idle CPU core into an unallocated group (U group); executing a training task by using a CPU core of the training task-specific group and, when there is a request for an online inference task, executing the online inference task by using the idle CPU core of the U group; and when there is a request for a batch inference task, additionally executing the batch inference task by using at least one of the idle CPU core of the U group and an idle CPU core of the training task-specific group.