A method for displaying images similar to a selected image includes receiving, from a user, a selection of an anchor image, generating, using a machine learning model, an anchor embeddings set for the anchor image and respective candidate embeddings sets for a plurality of candidate images. The method also includes calculating a distance between the anchor embeddings set and each of the plurality of candidate embeddings sets and displaying at least one of the plurality of candidate images based on the calculated distance.

Systems and methods for noise agnostic federated learning are disclosed. A method may include a client computer program executed by an electronic device in a federated learning computer network comprising a plurality of clients: receiving, from a federated learning computer program, a data format having desirable noise characteristics; transforming a client data set comprising variable noise characteristics to the data format using a client generative adversarial network (GAN); generating client weights for the transformed client data set, wherein the client weights indicate features of the client data set; communicating the client weights to the federated learning computer program; receiving, from the federated learning computer program, adjusted weights, wherein the adjusted weights are based on the client weights and a plurality client weights received from the clients in the federated learning computer network; and updating the client weights for a client machine learning model using the adjusted weights.

Systems and methods for noise agnostic federated learning are disclosed. A method may include a client computer program executed by an electronic device in a federated learning computer network comprising a plurality of clients: receiving, from a federated learning computer program, a data format having desirable noise characteristics; transforming a client data set comprising variable noise characteristics to the data format using a client generative adversarial network (GAN); generating client weights for the transformed client data set, wherein the client weights indicate features of the client data set; communicating the client weights to the federated learning computer program; receiving, from the federated learning computer program, adjusted weights, wherein the adjusted weights are based on the client weights and a plurality client weights received from the clients in the federated learning computer network; and updating the client weights for a client machine learning model using the adjusted weights.

Systems, methods and aspects, and embodiments thereof relate to unsupervised or semi-supervised features learning using a quantum processor. To achieve unsupervised or semi-supervised features learning, the quantum processor is programmed to achieve Hierarchal Deep Learning (referred to as HDL) over one or more data sets. Systems and methods search for, parse, and detect maximally repeating patterns in one or more data sets or across data or data sets. Embodiments and aspects regard using sparse coding to detect maximally repeating patterns in or across data. Examples of sparse coding include L0 and L1 sparse coding. Some implementations may involve appending, incorporating or attaching labels to dictionary elements, or constituent elements of one or more dictionaries. There may be a logical association between label and the element labeled such that the process of unsupervised or semi-supervised feature learning spans both the elements and the incorporated, attached or appended label.

A seismic dataset and a task to be performed with the seismic dataset may be received. A representative seismic line representative of the seismic dataset may be generated. The representative seismic line may include pixel data representative of the seismic dataset. Based on the representative seismic line, the task may be performed. The task may include at least finding an analogous geological region by searching for an analogous seismic dataset existing in a seismic database by comparing the representative seismic line with the analogous seismic dataset's representative seismic line.

In one embodiment, a diagnosis support apparatus includes: an input circuit configured to acquire a first medical image; and processing circuitry configured to generate a second medical image from the first medical image in such a manner that information included in the second medical image is reduced from information included in the first medical image, extract auxiliary information from the first medical image, and perform inference of a disease by using the second medical image and the auxiliary information.

Methods and apparatus are disclosed for implementing machine learning data augmentation within the die of a non-volatile memory (NVM) apparatus using on-chip circuit components formed on or within the die. Some particular aspects relate to configuring under-the-array or next-to-the-array components of the die to generate augmented versions of images for use in training a Deep Learning Accelerator of an image recognition system by rotating, translating, skewing, cropping, etc., a set of initial training images obtained from a host device. Other aspects relate to configuring under-the-array or next-to-the-array components of the die to generate noise-augmented images by, for example, storing and then reading training images from worn regions of a NAND array to inject noise into the images.

An apparatus comprising memory configured to store data to be machine-recognized (710), and at least one processing core configured to run an adaptive boosting machine learning algorithm with the data, wherein a plurality of learning algorithms are applied, wherein a feature space is partitioned into bins, wherein a distortion function is applied to features of the feature space (720), and wherein a first derivative of the distortion function is not constant (730).

Methods, apparatuses and systems directed to pattern learning, recognition, and metrology. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In certain implementations, the present invention provides for methods and systems suitable for analyzing and recognizing patterns in biological signals such as multi-electrode array waveform data. In other implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes. In other implementations, the present invention provides methods and systems for dynamic learning of patterns in supervised and unsupervised manners.

Provided are a drug recognizing apparatus, a drug recognizing method, and a drug recognizing program capable of enhancing robustness of a master image in a case where a drug is recognized. The drug recognizing apparatus includes an illumination unit that illuminates a drug; an imaging unit that images the illuminated drug; a storage unit that stores a master image for each drug type; a drug position acquiring unit that acquires a position of the drug on the basis of a captured image obtained by the imaging unit; a master image generating unit that generates the master image from a drug area in the captured image; an updating determination unit that determines whether to update the master image on the basis of the position of the drug acquired by the drug position acquiring unit; and a registration unit that registers the master image in the storage unit in a case where it is determined that the master image is to be updated.