Koplar Interactive Systems International, L.L.C. - St. Louis MO
International Classification:
H04B 3/20 H04M 9/08
US Classification:
381 66, 37940606, 37940614, 370286
Abstract:
Methods and systems for echo modulation are described. In one embodiment, intensities of a plurality of values in multiple windows of an audio signal may be obtained. The windows may be subject to a periodic boundary condition. A plurality of echo values may be calculated for each of the respective windows. The audio signal may be altered in one or more of the windows using a windowing function and echo values. Additional methods and systems are disclosed.
Method And System For Training A Neural Network For Generating Universal Adversarial Perturbations
- Cambridge MA, US Shuchin Aeron - Newton MA, US Adnan Rakin - Tempe AZ, US Toshiaki Koike Akino - Belmont MA, US Pierre Moulin - Urbana IL, US Kieran Parsons - Rockport MA, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
G06N 3/08 G06N 3/04 G06K 9/62
Abstract:
Embodiments of the present disclosure disclose a method and a system for training a neural network for generating universal adversarial perturbations. The method includes collecting a plurality of data samples. Each of the plurality of data samples is identified by a label from a finite set of labels. The method includes training a probabilistic neural network for transforming the plurality of data samples into a corresponding plurality of perturbed data samples having a bounded probability of deviation from the plurality of data samples by maximizing a conditional entropy of the finite set of labels of the plurality of data samples conditioned on the plurality of perturbed data samples. The conditional entropy is unknown. The probabilistic neural network is trained based on an iterative estimation of a gradient of the unknown conditional entropy of labels. The method further includes generating the universal adversarial perturbations based on the trained probabilistic neural network.
System And Method For Detecting Adversarial Attacks
A linguistic system for transcribing an input, where the linguistic system comprises a processor configured to execute a neural network multiple times while varying weights of at least some nodes of the neural network to produce multiple transcriptions of the input. Further, determine a distribution of pairwise distances of the multiple transcriptions; determine a legitimacy of the input based on the distribution; and transcribe the input using stored weights of the nodes of the neural network when the input is determined as legitimate to produce a final transcription of the input.
Pierre Moulin is French historian, specializing in World War II, Nisei Japanese Americans, the ... Moulin was born in Bruyeres-in-Vosges on November 1, 1948. ...