Abstract: A method for training an automated learning system includes processing training input with a first neural network and processing the output of the first neural network with a second neural network. The input layer of the second neural network corresponding to the output layer of the first neural network. The output layer of the second neural network corresponding to the input layer of the first neural network. An objective function is determined using the output of the second neural network and a predetermined modification magnitude. The objective function is approximated using random Cauchy projections which are propagated through the second neural network.

Abstract: A machine learning system includes encoder and decoder networks. The machine learning system is configured to obtain input data, which includes sensor data and a radius of an p norm ball of admissible perturbations. Input bounding data is generated based on the input data. First bounding data and second bounding data are generated by respectively propagating the input bounding data on first and second outputs of the encoder network. Third bounding data is generated in association with a latent variable based on the first bounding data and the second bounding data. Fourth bounding data is generated by propagating the third bounding data on an output of the decoder network. A robustness certificate is established with respect to the input data by generating a lower bound of an evidence lower bound based on the first, second, third, and fourth bounding data.

Abstract: An energy meter is configured to determine component waveforms that form a measured waveform. The meter inputs the waveform into one or more entries of a data structure, each entry of the one or more entries of the data structure storing a weight value that is determined based at least in part on values of the data signatures representing the plurality of remote devices, each entry being connected to one or more other entries of the data structure. The meter, for each of the one or more entries, generates an output value by performing an arithmetic operation on the waveform stored at that entry, the arithmetic operation comprising a function of the weight value. The meter identifies, from among the data signatures, one or more particular data signatures that are represented in the waveform. The meter determines, based on the particular data signatures, an operational state of another device.

Abstract: In a method for training a first neural network a superposed classification is back-propagated through a second neural network. An output value of the second neural network is utilized to determine whether the input of the first neural network is adversarial.

Abstract: A machine learning system includes encoder and decoder networks. The machine learning system is configured to obtain input data, which includes sensor data and a radius of an p norm ball of admissible perturbations. Input bounding data is generated based on the input data. First bounding data and second bounding data are generated by respectively propagating the input bounding data on first and second outputs of the encoder network. Third bounding data is generated in association with a latent variable based on the first bounding data and the second bounding data. Fourth bounding data is generated by propagating the third bounding data on an output of the decoder network. A robustness certificate is established with respect to the input data by generating a lower bound of an evidence lower bound based on the first, second, third, and fourth bounding data.

Abstract: A method for training an automated learning system includes processing training input with a first neural network and processing the output of the first neural network with a second neural network. The input layer of the second neural network corresponding to the output layer of the first neural network. The output layer of the second neural network corresponding to the input layer of the first neural network. An objective function is determined using the output of the second neural network and a predetermined modification magnitude. The objective function is approximated using random Cauchy projections which are propagated through the second neural network.

Abstract: In a method for training a first neural network a superposed classification is back-propagated through a second neural network. An output value of the second neural network is utilized to determine whether the input of the first neural network is adversarial.

Abstract: An energy meter is configured to determine component waveforms that form a measured waveform. The meter inputs the waveform into one or more entries of a data structure, each entry of the one or more entries of the data structure storing a weight value that is determined based at least in part on values of the data signatures representing the plurality of remote devices, each entry being connected to one or more other entries of the data structure. The meter, for each of the one or more entries, generates an output value by performing an arithmetic operation on the waveform stored at that entry, the arithmetic operation comprising a function of the weight value. The meter identifies, from among the data signatures, one or more particular data signatures that are represented in the waveform. The meter determines, based on the particular data signatures, an operational state of another device.