A computer system trains a federated learning model. A federated learning model is distributed to a plurality of computing nodes, each having a set of local training data comprising labeled data samples. Statistical data is received from each computing node that indicates the node's count of data samples for each label, and is analyzed to identify one or more computing nodes having local training data in which a label category is underrepresented beyond a threshold value with respect to data samples. Additional data samples labeled with the underrepresented labels are provided, and the computing nodes perform training. Results of training are received and are processed to generate a trained global model. Embodiments of the present invention further include a method and program product for training a federated learning model in substantially the same manner described above.

Systems and methods are provided for analyzing messages generated by a plurality of computing devices associated with a plurality of users in a messaging system to generate training data to train a machine learning model to determine a probability that a media content item was generated inside an enclosed location or outside, receiving a media content item from a computing device, analyzing the media content item using the trained machine learning model to determine a probability that the media content item was generated inside an enclosed location or outside, determining, based on the probability generated by the trained machine learning model, that the media content item was generated inside an enclosed location, and determining an inside temperature associated with the venue based on messages generated by a plurality of computing devices in a messaging system comprising media content items and temperature information for the venue or a similar venue type.

Systems, computer program products, and methods are described herein for preserving image and acoustic sensitivity using reinforcement learning. The present invention is configured to initiate a file editing engine on the audiovisual file to separate the audiovisual file into a video component and an audio component; initiate a convolutional neural network (CNN) algorithm on the video component to identify one or more sensitive portions in the one or more image frames; initiate an audio word2vec algorithm on the audio component to identify one or more sensitive portions in the audio component; initiate a masking algorithm on the one or more image frames and the audio component; generate a masked video component and a masked audio component based on at least implementing the masking action policy; and bind, using the file editing engine, the masked video component and the masked audio component to generate a masked audiovisual file.

Provided are methods for semantic annotation of sensor data using unreliable map annotation inputs, which can include training a machine learning model to accept inputs including images representing sensor data for a geographic area and unreliable semantic annotations for the geographic area. The machine learning model can be trained against validated semantic annotations for the geographic area, such that subsequent to training, additional images representing sensor data and additional unreliable semantic annotations can be passed through the neural network to provide predicted semantic annotations for the additional images. Systems and computer program products are also provided.

A method and system are provided which facilitate construction of an ensemble of neural network kernel classifiers. The system divides a training set into partitions. The system trains, based on the training set, a first neural network encoder to output a first set of features, and trains, based on each respective partition of the training set, a second neural network encoder to output a second set of features. The system generates, for each respective partition, based on the first and second set of features, kernel models which output a third set of features. The system classifies, by a classification model, the training set based on the third set of features. The generated kernel models for each respective partition and the classification model comprise the ensemble of neural network kernel classifiers. The system predicts a result for a testing data object based on the ensemble of neural network kernel classifiers.

A method for analyzing a specification parameter of an electronic component includes inputting a package type and at least one engineering drawing image of an electronic component; acquiring a probability value that in each view of the different viewing directions each of the plurality of specification parameter of the electronic component is labeled; taking the view of each of the plurality of specification parameters in the view direction with a highest probability value as a recommended view; performing a box selection on the plurality of specification parameters for at least one engineering drawing image with the same viewing direction as that of the recommended view by an object detection model; and identifying box-selected specification parameters to acquire a size value of identified specification parameters from the at least one engineering drawing image, and converting the size value into a corresponding editable text for output.

Techniques for building a machine learning framework with tracking, model building and maintenance, and feedback loop are provided. In one technique, a prediction model is generated based on features of multiple entities. For each entity indicated in a first database, multiple feature values are identified, which include feature values stored in the first database and feature values based on sub-entity data regarding individuals associated with the entity. The feature values are input into the prediction model to generate a score for the entity. Based on the score, a determination is made whether to add, to a second database, a record for that entity. The second database is analyzed to identify other entities. For each such entity, a determination is made whether to generate a training instance; if so, a training instance is generated and added to training data, which is used to generate another prediction model.

A technique is provided for training a prediction apparatus. The apparatus has an input interface for receiving a sequence of training events indicative of program instructions, and identifier value generation circuitry for performing an identifier value generation function to generate, for a given training event received at the input interface, an identifier value for that given training event. The identifier value generation function is arranged such that the generated identifier value is dependent on at least one register referenced by a program instruction indicated by that given training event. Prediction storage is provided with a plurality of training entries, where each training entry is allocated an identifier value as generated by the identifier value generation function, and is used to maintain training data derived from training events having that allocated identifier value. Matching circuitry is then responsive to the given training event to detect whether the prediction storage has a matching training entry (i.e. an entry whose allocated identifier value matches the identifier value for the given training event). If so, it causes the training data in the matching training entry to be updated in dependence on the given training event.

A computer-implemented method for reliably identifying objects in a sequence of input images received with the aid of an imaging sensor, positions of light sources in the respective input image being ascertained from the input images in each case with the aid of a first machine learning system, in particular, an artificial neural network, and objects from the sequence of input images being identified from the resulting sequence of positions of light sources, in particular, with the aid of a second machine learning system, in particular, with the aid of an artificial neural network.

Methods, systems, an apparatus, including computer programs encoded on a storage device, for training an image classifier. A method includes receiving an image that includes a depiction of an object; generating a set of poorly localized bounding boxes; and generating a set of accurately localized bounding boxes. The method includes training, at a first learning rate and using the poorly localized bounding boxes, an object classifier to classify the object; and training, at a second learning rate that is lower than the first learning rate, and using the accurately localized bounding boxes, the object classifier to classify the object. The method includes receiving a second image that includes a depiction of an object; and providing, to the trained object classifier, the second image. The method includes receiving an indication that the object classifier classified the object in the second image; and performing one or more actions.