A method for determining hard example sensor data inputs for training a task neural network is described. The task neural network is configured to receive a sensor data input and to generate a respective output for the sensor data input to perform a machine learning task. The method includes: receiving one or more sensor data inputs depicting a same scene of an environment, wherein the one or more sensor data inputs are taken during a predetermined time period; generating a plurality of predictions about a characteristic of an object of the scene; determining a level of inconsistency between the plurality of predictions; determining that the level of inconsistency exceeds a threshold level; and in response to the determining that the level of inconsistency exceeds a threshold level, determining that the one or more sensor data inputs comprise a hard example sensor data input.

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.

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.

A method of generating a pre-training model, an electronic device and a storage medium, which relate to a field of an artificial intelligence technology, in particular to a computer vision and deep learning technology. The method includes: determining a performance index set corresponding to a candidate model structure set, the candidate model structure set is determined from a plurality of model structures included in a search space, and the search space is a super-network-based search space; determining, from the candidate model structure set, a target model structure corresponding to each chip according to the performance index set, each target model structure is a model structure meeting a performance index condition; and determining, for each chip, the target model structure corresponding to the chip as a pre-training model corresponding to the chip, the chip is configured to run the pre-training model corresponding to the chip.

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 system, method, and computer-readable medium for associating a person’s gestures with specific features of objects is disclosed. Using one or more image capture devices, a person’s gestures and the location of that person in an environment is determined. Using determined distances between the person and objects in the environment and scales associated with features of those objects, the list of specific features in the person’s field-of-view may be determined. Further, a facial expression of the person may be scored and that score associated with one or more specific features.

A method for stabilizing an image based on artificial intelligence includes acquiring tremor detection data with respect to the image, the tremor detection data acquired from two or more sensors; outputting stabilization data for compensating for an image shaking, the stabilization data outputted using an artificial neural network (ANN) model trained to output the stabilization data based on the tremor detection data; and compensating for the image shaking using the stabilization data. A camera module includes a lens; an image sensor to output an image captured through the lens; two or more sensors to output tremor detection data with respect to the image; a controller to output stabilization data based on the tremor detection data using an ANN model; and a stabilization unit to compensate for an image shaking using the stabilization data. The ANN model is trained to output the stabilization data based on the tremor detection data.

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.

In a method for processing microscope images, at least one microscope image is provided as input image for an image processing algorithm. An output image is created from the input image by means of the image processing algorithm. The creation of the output image comprises adding low-frequency components for representing solidity of image structures of the input image to the input image, wherein the low-frequency components at least depend on high-frequency components of these image structures and wherein high-frequency components are defined by a higher spatial frequency than low-frequency components. A corresponding computer program and microscope system are likewise described.

Disclosed herein are system, method, and computer program product embodiments for adapting machine learning models for use in additional applications. For example, feature extraction models are readily available for use in applications such as image detection. These feature extraction models can be used to label inputs (such as images) in conjunction with other deep neural network models. However, in adapting the feature extraction models to these uses, it becomes problematic to improve the quality of their results on target data sets, as these feature extraction models are large and resistant to retraining. Approaches disclosed herein include a transfer layer for providing fast retraining of machine learning models.