A method of improving robustness of a deep neural network (DNN), the method including: applying a coverage metric to a trained DNN based on a test set to determine test set adequacy; monitoring a performance of the trained DNN; based on the performance, applying new data to the trained DNN; applying a novelty metric to an output of the trained DNN based on the applied new data to identify a subset of the applied new data in response to determining whether new features are generated; and identifying the subset of the applied new data.

A computer implemented method for the automated analysis or use of data is implemented by a voice assistant. The method comprises the steps of: (a) storing in a memory a structured, machine-readable representation of data that conforms to a machine-readable language (‘machine representation’); the machine representation including representations of user speech or text input to a human/machine interface; and (b) automatically processing the machine representations to analyse the user speech or text input.

An asynchronous distributed generative adversarial network (AsynDGAN) can include a central computing system and at least two discriminator nodes. The central computing system can include a generator neural network, an aggregator, and a network interface. Each discriminator node can have its own corresponding training data set. In addition, different discriminator nodes can use different data modalities. The central computing system communicates with each of the at least two discriminator nodes via the network interface and aggregates data received from the at least two discriminator nodes, via the aggregator, to update a model for the generator neural network during training of the generator neural network. The central computing system can further include a data access system that supports third party access to synthetic data generated by the generator neural network.

An asynchronous distributed generative adversarial network (AsynDGAN) can include a central computing system and at least two discriminator nodes. The central computing system can include a generator neural network, an aggregator, and a network interface. Each discriminator node can have its own corresponding training data set. In addition, different discriminator nodes can use different data modalities. The central computing system communicates with each of the at least two discriminator nodes via the network interface and aggregates data received from the at least two discriminator nodes, via the aggregator, to update a model for the generator neural network during training of the generator neural network. The central computing system can further include a data access system that supports third party access to synthetic data generated by the generator neural network.

A system for controlling multicomputer interaction with deep learning is disclosed that includes a controller system that is configured to generate one or more first user-controllable avatars on an interaction field, where the first avatars include movement controls and prompt functionality that is controllable by a first user to cause the first avatar to generate a prompt. A client system is configured to generate a second user-controllable avatar on the interaction field, where the second avatar includes movement controls and response functionality that is controllable by a second user to cause the second avatar to generate a response to the prompt. A deep learning processing system is configured to receive the prompt and the response and to process the prompt and the response to generate a score and to assign the score to one of two or more categories associated with the second user.

Techniques for configuring one or more applications based on a detected wakefulness state of a user are disclosed. A system trains and applies a machine learning model to wakefulness data to compute a wakefulness state of a user. The system obtains the wakefulness data from wearable devices worn by the user and environmental devices in a user's environment. The system configures applications and/or devices based on the computed wakefulness state of the user. The system configures the ability of devices or applications to generate visual, audible, or tactile notifications in response to determining that a user is awake or asleep.

The present specification discloses video matching. In a computer-implemented method, a plurality of feature vectors of a target video is obtained. A candidate video similar to the target video is retrieved from a video database based on the plurality of feature vectors of the target video. A time domain similarity matrix feature map is constructed between the target video and the candidate video based on the target video and the candidate video. Using the time domain similarity matrix feature map as an input into a deep learning detection model, a video segment matching the target video in the candidate video and a corresponding similarity is output.

The present specification discloses video matching. In a computer-implemented method, a plurality of feature vectors of a target video is obtained. A candidate video similar to the target video is retrieved from a video database based on the plurality of feature vectors of the target video. A time domain similarity matrix feature map is constructed between the target video and the candidate video based on the target video and the candidate video. Using the time domain similarity matrix feature map as an input into a deep learning detection model, a video segment matching the target video in the candidate video and a corresponding similarity is output.

A signal classification method with noise immunity, includes receiving a wireless training signal; combining the wireless training signal with a white Gaussian noise signal, based on a preset desired signal-to-noise ratio (SNR), to generate a modulation signal resulting from modulating an SNR of the wireless training signal to correspond to the preset desired SNR; generating, based on a signal resulting from performing short-time Fourier transform on the modulation signal, a power-based spectrogram image corresponding to the wireless training signal; inputting the power-based spectrogram image and a predetermined supervised learning value corresponding to the wireless training signal into a preset convolution neural network (CNN) model to train the preset CNN model; generating, based on a signal resulting from performing short-time Fourier transform on the wireless evaluation signal, a power-based spectrogram image corresponding to the wireless evaluation signal; and classifying the wireless evaluation signal by applying the power-based spectrogram image.

A signal classification method with noise immunity, includes receiving a wireless training signal; combining the wireless training signal with a white Gaussian noise signal, based on a preset desired signal-to-noise ratio (SNR), to generate a modulation signal resulting from modulating an SNR of the wireless training signal to correspond to the preset desired SNR; generating, based on a signal resulting from performing short-time Fourier transform on the modulation signal, a power-based spectrogram image corresponding to the wireless training signal; inputting the power-based spectrogram image and a predetermined supervised learning value corresponding to the wireless training signal into a preset convolution neural network (CNN) model to train the preset CNN model; generating, based on a signal resulting from performing short-time Fourier transform on the wireless evaluation signal, a power-based spectrogram image corresponding to the wireless evaluation signal; and classifying the wireless evaluation signal by applying the power-based spectrogram image.