A method for training an object detection network by using attention maps is provided. The method includes steps of: (a) an on-device learning device inputting the training images into a feature extraction network, inputting outputs of the feature extraction network into a attention network and a concatenation layer, and inputting outputs of the attention network into the concatenation layer; (b) the on-device learning device inputting outputs of the concatenation layer into an RPN and an ROI pooling layer, inputting outputs of the RPN into a binary convertor and the ROI pooling layer, and inputting outputs of the ROI pooling layer into a detection network and thus to output object detection data; and (c) the on-device learning device train at least one of the feature extraction network, the detection network, the RPN and the attention network through backpropagations using an object detection losses, an RPN losses, and a cross-entropy losses.

A system to implement a deep learning network model is disclosed. The system includes machine-readable instructions and data that include a biomedical information library comprising information that includes a plurality of drugs, a plurality of biological targets, a plurality of diseases, and a plurality of adverse effects, a biomedical network system comprising a plurality of networks covering chemical, genomic, phenotypic, and cellular profiles, a score prioritizer to determine new targets for the plurality of drugs based on a concatenation of a low-dimensional vector representation for each drug vertex and each biological target vertex, and a model generator configured to generate a deep learning network model that defines a plurality of relationships between the drugs, the biological targets, and the adverse effects. The plurality of relationships defined by the deep learning network model predict drug target identification and drug repurposing.

A bi-directional spatial-temporal transformer neural network (BDSTT) is trained to predict original coordinates of a skeletal joint in a specific frame through relative relationships of the skeletal joint to other joints and to the state of the skeletal joint in other frames. Obtain a plurality of frames comprising coordinates of the skeletal joint and coordinates of other joints. Produce a spatially masked frame by masking the original coordinates of the skeletal joint. Provide the specific frame, the spatially masked frame, and at least one more frame to a coordinate prediction head of the BDSTT. Obtain, from the coordinate prediction head, a prediction of coordinates for the skeletal joint. Adjust parameters of the BDSTT until a mean-squared error, between the prediction of coordinates for the skeletal joint and the original coordinates of the skeletal joint, converges.

A bi-directional spatial-temporal transformer neural network (BDSTT) is trained to predict original coordinates of a skeletal joint in a specific frame through relative relationships of the skeletal joint to other joints and to the state of the skeletal joint in other frames. Obtain a plurality of frames comprising coordinates of the skeletal joint and coordinates of other joints. Produce a spatially masked frame by masking the original coordinates of the skeletal joint. Provide the specific frame, the spatially masked frame, and at least one more frame to a coordinate prediction head of the BDSTT. Obtain, from the coordinate prediction head, a prediction of coordinates for the skeletal joint. Adjust parameters of the BDSTT until a mean-squared error, between the prediction of coordinates for the skeletal joint and the original coordinates of the skeletal joint, converges.

Generating soft labels used for training a unified model is achieved by unification of models having respective target classes with distillation. A collection of samples is prepared. Predictions are generated by individual trained models. Individual trained models have an individual class set to form a unified class set that includes target classes. The unified soft labels are estimated for each sample over the target classes in the unified class set from the predictions using a relation connecting a first output of each individual trained model and a second output of the unified model. The unified soft labels are output to train a unified model having the unified class set.

A measurement apparatus comprising an acquisition memory adapted to store data sections of at least one acquired measurement signal; a processor adapted to calculate a measurement parameter vector, v, for each data section of the acquired measurement signal; and a trained autoencoder neural network adapted to process the measurement parameter vectors, v, applied as input data to the trained autoencoder neural network to provide at a middle layer of said autoencoder neural network an encoded vector, h, with characteristic signal features of the acquired measurement signal.

Some techniques described herein relate to generating a hierarchical topic model (HTM), which can be used to generate custom content. In one example, a method includes determining first-level topics in a topic hierarchy related to a corpus of documents. A first-level topic of the first-level topics includes multiple words. The multiple words are grouped into clusters based on word embeddings of the multiple words. The multiple words are then subdivided into second-level topics as subtopics of the first-level topic, such that the number of second-level topics equals the number of clusters. A document of the corpus of documents is assigned to the first-level topic and to a second-level topic of the second-level topics, and an indication is received of access by a user to the document. Custom content is generated for the user based on one or more other documents assigned to the first-level topic and the second-level topic.

Some techniques described herein relate to generating a hierarchical topic model (HTM), which can be used to generate custom content. In one example, a method includes determining first-level topics in a topic hierarchy related to a corpus of documents. A first-level topic of the first-level topics includes multiple words. The multiple words are grouped into clusters based on word embeddings of the multiple words. The multiple words are then subdivided into second-level topics as subtopics of the first-level topic, such that the number of second-level topics equals the number of clusters. A document of the corpus of documents is assigned to the first-level topic and to a second-level topic of the second-level topics, and an indication is received of access by a user to the document. Custom content is generated for the user based on one or more other documents assigned to the first-level topic and the second-level topic.

Machine-learning methods and apparatus are provided to solve blind source separation problems with an unknown number of sources and having a signal propagation model with features such as wave-like propagation, medium-dependent velocity, attenuation, diffusion, and/or advection, between sources and sensors. In exemplary embodiments, multiple trials of non-negative matrix factorization are performed for a fixed number of sources, with selection criteria applied to determine successful trials. A semi-supervised clustering procedure is applied to trial results, and the clustering results are evaluated for robustness using measures for reconstruction quality and cluster separation. The number of sources is determined by comparing these measures for different trial numbers of sources. Source locations and parameters of the signal propagation model can also be determined. Disclosed methods are applicable to a wide range of spatial problems including chemical dispersal, pressure transients, and electromagnetic signals, and also to non-spatial problems such as cancer mutation.

A computing system can include an embedding model and a clustering model. The computing system input each of the plurality of inputs into the embedding model and receiving respective embeddings for the plurality of inputs as outputs of the embedding model. The computing system can input the respective embeddings for the plurality of inputs into the clustering model and receiving respective cluster assignments for the plurality of inputs as outputs of the clustering model. The computing system can evaluate a clustering loss function that evaluates a first average, across the plurality of inputs, of a respective first entropy of each respective probability distribution; and a second entropy of a second average of the probability distributions for the plurality of inputs. The computing system can modify parameter(s) of one or both of the clustering model and the embedding model based on the clustering loss function.