According to an embodiment, a processing apparatus includes a hardware processor. The hardware processor is configured to: cut out, from an input signal, a plurality of partial signals that are predetermined parts in the input signal; execute processing on the plurality of partial signals using neural networks having the same layer structure with each other to generate a plurality of intermediate signals including a plurality of signals corresponding to a plurality of channels; execute predetermined statistical processing on signals for each of the plurality of channels for each of the plurality of intermediate signals corresponding to the plurality of partial signals, to calculate statistics for each channel and generate a concatenated signal by concatenating the statistics of the plurality of respective intermediate signals for each channel; generate a synthetic signal by performing predetermined processing on the concatenated signal; and output an output signal in accordance with the synthetic signal.

A prediction unit classifies input data into a plurality of classes using a predictive model, and outputs a predicted probability for each class as a prediction result. A grouping unit generates a grouped class formed by k classes within top k predicted probabilities, and calculates a predicted probability of the grouped class. A loss calculation unit calculates a loss based on predicted probabilities of a plurality of classes including the grouped class. A model update unit updates the predictive model based on the calculated loss.

A method may comprise receiving, via a processor, a first indication that an object is in a first zone of interest of a first sensor in the plurality of sensors; receiving, via the processor, a second indication that the object is in a second zone of interest of a second sensor in the plurality of sensors; and determining, via the processor, whether the first sensor or the second sensor is falsely detecting the object within the respective zone of interest.

In real biometric systems, false match rates and false non-match rates of 0% do not exist. There is always some probability that a purported match is false, and that a genuine match is not identified. The performance of biometric systems is often expressed in part in terms of their false match rate and false non-match rate, with the equal error rate being when the two are equal. There is a tradeoff between the FMR and FNMR in biometric systems which can be adjusted by changing a matching threshold. This matching threshold can be automatically, dynamically and/or user adjusted so that a biometric system of interest can achieve a desired FMR and FNMR.

Techniques for extending sensitive data tagging without reannotating training data are described. A method for extending sensitive data tagging without reannotating training data may include hosting a plurality of models at a model endpoint in a machine learning service, each model trained to identify a different sensitive data type in a transcript of content, adding a new model to the model endpoint, the new model trained to identify a new sensitive data entity in the transcript of content, identifying sensitive entities in the transcript by each of the plurality of models and the new model, merging inference responses generated by each of the plurality of models and the new model using at least one inference policy, and returning a merged inference response identifying a plurality of sensitive entities in the transcript.

An information processing device includes a processor. The processor obtains an input image, inputs the input image to a machine learning model that executes classification likelihood calculation processing to obtain, for each of candidate objects in the input image, likelihoods of belonging to the plurality of classes, executes first determination on whether or not each of the candidate objects is classified as a first class of the plurality of classes using a likelihood of belonging to the first class that is a likelihood having a negative correlation with likelihoods of belonging to other classes, executes second determination on whether or not each of the candidate objects that have been determined in the first determination as a non-first class is classified as the other classes, and outputting a result of classifying the candidate objects included in the input image using a result of the second determination.

A method and apparatus for obtaining word vectors based on a language model, a device and a storage medium are disclosed, which relates to the field of natural language processing technologies in artificial intelligence. An implementation includes inputting each of at least two first sample text language materials into the language model, and outputting a context vector of a first word mask in each first sample text language material via the language model; determining the word vector corresponding to each first word mask based on a first word vector parameter matrix, a second word vector parameter matrix and a fully connected matrix respectively; and training the language model and the fully connected matrix based on the word vectors corresponding to the first word masks in the at least two first sample text language materials, so as to obtain the word vectors.

A method may include applying, to a corpus of data, a first machine learning technique to identify candidate domains of an ontology mapping brain structure to mental function. The corpus of data may include textual data describing a plurality of mental functions and spatial data corresponding to a plurality of brain structures. A second machine technique may be applied to optimize a quantity of domains included in the ontology and/or a quantity of mental function terms included in each domain. The ontology may be applied to phenotype an electronic medical record and predict a clinical outcome for a patient associated with the electronic medical record. Related systems and articles of manufacture, including computer program products, are also provided.

Embodiments determine anomalies in sensor data generated by a sensor by receiving an evaluation time window of clean sensor data generated by the sensor. Embodiments receive a threshold value for determining anomalies. When the clean sensor data has a cyclic pattern, embodiments divide the evaluation time window into a plurality of segments of equal length, wherein each equal length comprises the cyclic pattern. When the clean sensor data does not have the cyclic pattern, embodiments divide the evaluation time window into a pre-defined number of plurality of segments of equal length. Embodiments convert the evaluation time window and each of the plurality of segments into corresponding curves using Kernel Density Estimation (“KDE”). For each of the plurality of segments, embodiments determine a Kullback-Leibler (“KL”) divergence value between corresponding curves of the segment and the evaluation time window to generate a plurality of KL divergence values.

Certain aspects involve evaluating modeling algorithms whose outputs can impact machine-implemented operating environments. For instance, a computing system generates, from a comparison of a set of estimated attribute values of an attribute to a set of validation attribute values of the attribute, a discretized evaluation dataset with data values in multiple categories. The computing system computes, for a modeling algorithm used to generate the estimated attribute values, an evaluation metric. The computing system provides a host computing system with access to the evaluation metric, one or more modeling outputs generated with the modeling algorithm, or both. Providing one or more of these outputs to the host computing system can facilitate modifying one or more machine-implemented operations.