Abstract: The present invention relates to methods and systems for implicitly computing pairwise sequence attention scores with sub-quadratic complexity in artificial neural networks. More specifically, the present application discloses an “implicit attention” mechanism that computes a pairwise attention score on the output of a neural network at each step in an input sequence, rather than across these steps. This implicit attention mechanism operates by taking the output vector produced by a neural network layer for a given sequence step, reshaping this vector into a matrix, and then transforming this matrix on the basis of pairwise similarities between its rows and columns, so as to produce an output vector that stores a compressed summary of all of the sequential dependencies present in the input sequence that are relevant for performing at least one classification, regression, or data generation task.

Abstract: An example process includes: obtaining information about call-outs from at least one bidder of an online content auction, where the information spans a time period and the information identifies which call-outs, within the time period, did not result in a response; identifying, based on the information, one or more features that are common to call-outs that did not result in at least a threshold number of responses from the at least one bidder; generating a model for the at least one bidder that predicts whether the at least one bidder will respond to a call-out associated with the one or more features; using the model to determine a percentage of additional call-outs having the one or more features to send to the at least one bidder; and sending the percentage of additional call-outs to the at least one bidder.

Abstract: An example process includes: obtaining information about call-outs from at least one bidder of an online content auction, where the information spans a time period and the information identifies which call-outs, within the time period, did not result in a response; identifying, based on the information, one or more features that are common to call-outs that did not result in at least a threshold number of responses from the at least one bidder; generating a model for the at least one bidder that predicts whether the at least one bidder will respond to a call-out associated with the one or more features; using the model to determine a percentage of additional call-outs having the one or more features to send to the at least one bidder; and sending the percentage of additional call-outs to the at least one bidder.