Abstract: In some embodiments, noise data may be used to train a neural network (or other prediction model). In some embodiments, input noise data may be obtained and provided to a prediction model to obtain an output related to the input noise data (e.g., the output being a prediction related to the input noise data). One or more target output indications may be provided as reference feedback to the prediction model to update one or more portions of the prediction model, wherein the one or more portions of the prediction model are updated based on the related output and the target indications. Subsequent to the portions of the prediction model being updated, a data item may be provided to the prediction model to obtain a prediction related to the data item (e.g., a different version of the data item, a location of an aspect in the data item, etc.).

Abstract: In some embodiments, anomaly detection may be facilitated via a multi-neural-network architecture. In some embodiments, a first neural network may be configured to generate hidden representations of data items corresponding to a concept. A second neural network may be configured to generate reconstructions of the data items from the hidden representations. The first neural network may be configured to assess the reconstructions against the data items and update configurations of the first neural network based on the assessment of the reconstructions. Subsequent to the update of the first neural network, the first neural network may generate a hidden representation of a first data item from the first data item. The second neural network may generate a reconstruction of the first data item from the hidden representation. An anomaly in the first data item may be detected based on differences between the first data item and the reconstruction.