Various embodiments are provided for providing machine learning with noisy label data in a computing environment using one or more processors in a computing system. A label corruption probability of noisy labels may be estimated for selected data from a dataset using temporal inconsistency in a machine model prediction during a training operation in a neural network.

Apparatus and associated methods relate to a pattern recognition system configured to classify a transaction as anomalous or not anomalous as a function of a predictive analytic model configured to detect anomalies, generate a rule based on expert analysis of a limited number of data samples to classify as anomalous a transaction erroneously classified as not anomalous, augment the predictive analytic model with the generated rule, and deploy the augmented predictive analytic model to automatically identify an attack early in a live transaction stream. In some examples, the transaction may be a bank card purchase. Some transactions may be classified anomalous due to fraud, compliance violation such as money laundering, or terrorist funding. The predictive analytic model may be, for example, a decision tree followed by a regression model. Various embodiments may advantageously generate rules based on transaction criteria selected by human experts exploring and manipulating visually perceptible transaction representations.

Techniques are provided for training machine learning model. According to one aspect, a training data is received by one or more processing units. The machine learning model is trained based on the training data, wherein the training comprises: optimizing the machine learning model based on stochastic gradient descent (SGD) by adding a dynamic noise to a gradient of a model parameter of the machine learning model calculated by the SGD.

This disclosure relates to improved techniques for performing computer vision functions on medical images, including object segmentation functions for identifying medical objects in the medical images and grading functions for determining severity labels for medical conditions exhibited in the medical images. The techniques described herein utilize a neural network architecture to perform these and other functions. The neural network architecture can be trained, at least in part, using semi-supervised learning techniques that enable the neural network architecture to accurately perform the object segmentation and grading functions despite limited availability of pixel-level annotation information.

Disclosed are methods, systems, and other implementations for processing, analyzing, and modelling psychotherapy data. The implementations include a method for analyzing psychotherapy data that includes obtaining transcript data representative of spoken dialog in one or more psychotherapy sessions conducted between a patient and a therapist, extracting speech segments from the transcript data related to one or more of the patient or the therapist, applying a trained machine learning topic model process to the extracted speech segments to determine weighted topic labels representative of semantic psychiatric content of the extracted speech segments, and processing the weighted topic labels to derive a psychiatric assessment for the patient.