Techniques are provided for deep neural network (DNN) identification of realistic synthetic images generated using a generative adversarial network (GAN). According to an embodiment, a system is described that can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise, a first extraction component that extracts a subset of synthetic images classified as non-real like as opposed to real-like, wherein the subset of synthetic images were generated using a GAN model. The computer executable components can further comprise a training component that employs the subset of synthetic images and real images to train a DNN network model to classify synthetic images generated using the GAN model as either real-like or non-real like.

The technology disclosed presents a system that comprises a memory, a data partitioning logic, and an annotation logic. The memory stores a sequence of frames of a video. The data partitioning logic is configured to partition the sequence of frames into an oracle set and an unannotated set. Frames in the oracle set are annotated by a user. Frames in the unannotated set are candidates for user annotation conditional upon being members of a core set, and for machine annotation conditional upon being non-members of the core set. The annotation logic is configured to generate annotations for the frames in the unannotated set. The annotations include user annotations based on membership in the core set, and machine annotations based on non-membership in the core set.

Intent determination based on one or more multi-model structures can include generating an output from each of a plurality of domain-specific models in response to a received input. The domain-specific models can comprise simultaneously trained machine learning models that are trained using a corresponding local loss metric for each domain-specific model and a global loss metric for the plurality of domain-specific models. The presence or absence of an intent corresponding to one or more domain-specific models can be determined by classifying the output of each domain-specific model.

The technology disclosed assigns quality scores to bases called by a neural network-based base caller by (i) quantizing classification scores of predicted base calls produced by the neural network-based base caller in response to processing training data during training, (ii) selecting a set of quantized classification scores, (iii) for each quantized classification score in the set, determining a base calling error rate by comparing its predicted base calls to corresponding ground truth base calls, (iv) determining a fit between the quantized classification scores and their base calling error rates, and (v) correlating the quality scores to the quantized classification scores based on the fit.

Provided is a computer-implemented method for training a machine learning (ML) model using labelled and unlabelled data, the method comprising obtaining a set or training data comprising a set of labelled data items and a set of unlabelled data items, training a loss module of the ML model using labels in the set of labelled data items, to generate a trained loss module capable of estimating a likelihood of a label for a data item, and training a task module of the ML model using the loss module, the set of labelled data items, and the set of unlabelled data items, to generate a trained task module capable of making a prediction of a label for input data.

The system can include a plurality of data processing modules (e.g., machine learning models, rule-based models, etc.), which can include: a feature generation module, a Driver versus Passenger (DvP) classification module, a validation module, an update module, and/or any other suitable data processing modules. The system can optionally include a mobile device (e.g., such as a mobile cellular telephone, user device, etc.) and/or can be used in conjunction with a mobile device (e.g., receive data from an application executing at the mobile device and/or utilize mobile device processing, etc.). However, the system can additionally or alternatively include any other suitable set of components. The system functions to facilitate execution of method S100. Additionally or alternatively, the system functions to classify a role of a mobile device user (e.g., driver or passenger; driver or nondriver; etc.) for a vehicle trip and/or determine a semantic label for the vehicle trip.

A method for generating a neural network for detecting one or more objects in images includes generating one or more self-supervised proposal learning losses based on the one or more proposal features and corresponding proposal feature predictions. One or more consistency-based proposal learning losses are generated based on noisy proposal feature predictions and the corresponding proposal predictions without noise. A combined loss is generated using the one or more self-supervised proposal learning losses and one or more consistency-based proposal learning losses. The neural network is updated based on the combined loss.

A cascade auto-review system for automated classification and annotation of input is provided. An example system is structure adaptive and task oriented and includes a communication module configured to receive the input including images, videos, and metadata. The system further includes a plurality of subsystems. Each subsystem has a series of successive classifier stages configured to detect tags in the input and approve or reject the tags based on the images, the videos, and the metadata. The system further includes a database to store results of the classification and annotation. The results are used to train computer vision and machine learning algorithms.

Methods and systems for a data marketplace in a conveyor environment includes a self-organizing data marketplace. The self-organizing data marketplace includes at least one data collector and at least one corresponding conveyor in an industrial environment, wherein the at least one data collector is structured to collect detection values from at least one sensor of a power roller of the at least one corresponding conveyor; a data storage structured to store a data pool comprising at least a portion of the detection values; a data marketplace structured to self-organize the data pool; and a transaction system structured to interpret a user data request, and to selectively provide a portion of the self-organized data pool to a user in response to the user data request.

The present invention relates to a system that can be used to accurately classify objects in biological specimens. The user firstly classifies manually an initial set of images, which are used to train a classifier. The classifier then is run on a complete set of images, and outputs not merely the classification but the probability that each image is in a variety of classes. Images are then displayed, sorted not merely by the proposed class but also the likelihood that the image in fact belongs in a proposed alternative class. The user can then reclassify images as required.