Systems and methods for localization and classification of content in a data set

A device comprising:

• one or more processors configured to:
• provide a sequence of portions of a data set to a neural network to generate a plurality of activations, each activation of the plurality of activations comprising at least one value from a layer of the neural network;
• apply an attention vector to each activation of the plurality of activations to generate a sequence of values;
• generate, using a normalization function, a sequence of attention scores according to the sequence of values;
• and
• identify a first portion in the sequence of portions of the data set corresponding to a first attention score of the sequence of attention scores.

A non-transitory computer readable medium storing instructions when executed by one or more processors cause the one or more processors to:

• provide a sequence of portions of a data set to a neural network to generate a plurality of activations, each activation of the plurality of activations comprising at least one value from a layer of the neural network;
• apply an attention vector to each activation of the plurality of activations to generate a sequence of values;
• generate, using a normalization function, a sequence of attention scores according to the sequence of values;
• and
• identify a first portion in the sequence of portions of the data set corresponding to a first attention score of the sequence of attention scores.