Certain aspects of the present disclosure provide techniques for performing machine learning computations in a compute in memory (CIM) array comprising a plurality of bit cells, including: determining that a sparsity of input data to a machine learning model exceeds an input data sparsity threshold; disabling one or more bit cells in the CIM array based on the sparsity of the input data prior to processing the input data; processing the input data with bit cells not disabled in the CIM array to generate an output value; applying a compensation to the output value based on the sparsity to generate a compensated output value; and outputting the compensated output value.

The presently disclosed subject matter includes a computerized method and system that provide the ability to train and execute a unique machine learning (ML) model specifically configured to enhance classifier (e.g., RegEx) output by identifying and removing false positive results from the classifiers output. Classifier output, comprising a collection of data-subsets (e.g., columns in a relational database) of one or more structured or semi-structured data sources (e.g., tables of a relational database), are transformed to be represented by a plurality of numerical vectors. The numerical vectors are used during a training phase (as well as the execution phase) for training a machine learning model to enhance the classifier output and reduce false positives.

Methods, systems and apparatus, including computer programs encoded on computer storage medium, for training a neurosymbolic data imputation system on training data inputs in a domain to impute missing data in a data input from the data domain. In one aspect a method includes, for each training data input, adding random noise to missing fields of the training data input;

generating an embedding data input for the training data input using concept embeddings from the domain; processing the noisy data input and the embedding data input through a correlation network to obtain correlation data; applying attention to the noisy training data input and the correlation data to generate a combined data input; processing, by an autoencoder, the combined data input to obtain a decoded data output; computing a difference between the decoded data output and the training data input; and updating parameters of the data imputation system using the difference.

Systems and methods disclosed herein use a first machine learning architecture and a second machine learning architecture where the first machine learning architecture executes on a first processor and receives a first image representing a mouth of a user, determines user feedback for outputting to the user based on a first machine learning model, and outputs the user feedback for capturing a second image representing the mouth of the user. The second machine learning architecture executes on a second processor and receives the first image and the second image, and generates a 3D model of at least a portion of a dental arch of the user based on the first image and the second image where the 3D model is generated based on a second machine learning model of the second machine learning architecture.

Techniques for implementing a performance-adaptive sampling strategy towards fast and accurate graph neural networks are provided. In one technique, a graph that comprises multiple nodes and edges connecting the nodes is stored. An embedding for each node is initialized, as well as a sampling policy for sampling neighbors of nodes. One or more machine learning techniques are used to train a graph neural network and learn embeddings for the nodes. Using the one or more machine learning techniques comprises, for each node: (1) selecting, based on the sampling policy, a set of neighbors of the node; (2) based on the graph neural network and embeddings for the node and the set of neighbors, computing a performance loss; and (3) based on a gradient of the performance loss, modifying the sampling policy.

Various methods and systems are provided for computed tomography imaging. In one embodiment, a method includes acquiring, with an x-ray detector and an x-ray source coupled to a gantry, a three-dimensional image volume of a subject while the subject moves through a bore of the gantry and the gantry rotates the x-ray detector and the x-ray source around the subject, inputting the three-dimensional image volume to a trained deep neural network to generate a corrected three-dimensional image volume with a reduction in aliasing artifacts present in the three-dimensional image volume, and outputting the corrected three-dimensional image volume. In this way, aliasing artifacts caused by sub-sampling may be removed from computed tomography images while preserving details, texture, and sharpness in the computed tomography images.

A method for predicting a structure of an indoor space using radio propagation channel analysis through deep-learning is disclosed. Channel data of radio signals are collected for various indoor spaces, and radio channel parameter data such as PDP, AoA, and AoD are extracted therefrom. A large amount of propagation channel parameter data is input to an artificial neural network together with vertex coordinate data of the corresponding indoor space and deep-learning is performed in advance. The propagation channel parameter data are extracted from the indoor space to be predicted, the best matching indoor space is detected based on the trained artificial neural network. The best matching indoor space is predicted as the structure of the indoor space.

A generative system for the creation of digital images for printing on design surfaces comprises a training dataset comprising a plurality of sample images for printing on design surfaces, a generative adversarial network comprising a generator and a discriminator, wherein the generator receives noise at input and is trained to generate at output starting from the noise a new artificially generated image adapted to be used for printing on design surfaces, and wherein the discriminator receives at input the new artificially generated image and is trained to compare and distinguish the new image generated by the sample images of the training dataset.

The present invention is a molecular graph representation learning method based on contrastive learning, the method comprising: obtaining a molecular fingerprint representation of each molecule, and calculating a similarity between each two molecular fingerprints; collecting a full amount of chemical functional group information, and matching a corresponding functional group for each atom in the molecule; using a heterogeneous graph to model a molecular graph; using a RGCN in the structure-aware molecular encoder to encode the representation of each atom in the molecule and the representation of the functional group to which the atom belongs, and mapping the molecule to a feature space through an aggregation function to obtain a structure-aware feature representation; according to the fingerprint similarity between molecules, selecting positive and negative samples, and carrying out a comparative learning in the feature space; obtaining the structure-aware molecular encoder by using the contrastive learning method for training on a large-sample molecular dataset, and applying the structure-aware molecular encoder to a prediction task of downstream molecular attributes. The present invention helps to capture more abundant molecular structure information and solve the problem on molecular property prediction.

A method for training a machine learning model using information pertaining to characteristics of upload activity performed at one or more client devices includes generating first training input including (i) information identifying first amounts of data uploaded during a specified time interval for one or more of multiple application categories, and (ii) information identifying first locations external to a client device to which the first amounts of data are uploaded. The method includes generating a first target output that indicates whether the first amounts of data uploaded to the first locations correspond to malicious or non-malicious upload activity. The method includes providing the training data to train the machine learning model on (i) a set of training inputs including the first training input, and (ii) a set of target outputs including the first target output.