Abstract: Example implementations include methods, apparatuses, and computer-readable mediums of federated learning by a federated client device, comprising identifying client invariant information of a neural network for performing a machine learning (ML) task in a first domain known to a federated server. The implementations further comprising transmitting the client invariant information to the federated server, the federated server configured to generate a ML model for performing the ML task in a domain unknown to the federated server based on the client invariant information and other client invariant information of another neural network for performing the ML task in a second domain known to the federated server.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for a training and using machine learning models in multi-device network environments. An example computer-implemented method for network communications performed by a host device includes extracting a feature set from a data set associated with a client device using a client-device-specific feature extractor, wherein the feature set comprises a subset of features in a common feature space, training a task-specific model based on the extracted feature set and one or more other feature sets associated with other client devices, wherein the feature sets associated with the other client devices comprise one or more subsets of features in the common feature space, and deploying, to each respective client device of a plurality of client devices, a respective version of the task-specific model.

Abstract: Systems and techniques are provided for training a neural network model or machine learning model. For example, a method of augmenting training data can include augmenting, based on a randomly initialized neural network, training data to generate augmented training data and aggregating data with a plurality of styles from the augmented training data to generate aggregated training data. The method can further include applying semantic-aware style fusion to the aggregated training data to generate fused training data and adding the fused training data as fictitious samples to the training data to generate updated training data for training the neural network model or machine learning model.

Abstract: Imaging systems and techniques are described. An imaging system receives image data representing at least a portion (e.g., a face) of a first user as captured by a first image sensor. The imaging system identifies that a gaze of the first user as represented in the image data is directed toward a displayed representation of at least a portion (e.g., a face) of a second user. The imaging system identifies an arrangement of representations of users for output. The imaging system generates modified image data based on the gaze and the arrangement at least in part by modifying the image data to modify at least the portion of the first user in the image data to be visually directed toward a direction corresponding to the second user based on the gaze and the arrangement. The imaging system outputs the modified image data arranged according to the arrangement.

Abstract: Imaging systems and techniques are described. An imaging system receives image data representing at least a portion (e.g., a face) of a first user as captured by a first image sensor. The imaging system identifies that a gaze of the first user as represented in the image data is directed toward a displayed representation of at least a portion (e.g., a face) of a second user. The imaging system identifies an arrangement of representations of users for output. The imaging system generates modified image data based on the gaze and the arrangement at least in part by modifying the image data to modify at least the portion of the first user in the image data to be visually directed toward a direction corresponding to the second user based on the gaze and the arrangement. The imaging system outputs the modified image data arranged according to the arrangement.

Abstract: Techniques and apparatus for training a neural network to classify audio into one of a plurality of categories and using such a trained neural network. An example method generally includes receiving a data set including a plurality of audio samples. A relaxed feature-normalized data set is generated by normalizing each audio sample of the plurality of audio samples. A neural network is trained to classify audio into one of a plurality of categories based on the relaxed feature-normalized data set, and the trained neural network is deployed.

Abstract: Certain aspects of the present disclosure provide techniques for training a machine learning model. The method generally includes receiving, at a local device from a server, information defining a global version of a machine learning model. A local version of the machine learning model and a local center associated with the local version of the machine learning model are generated based on embeddings generated from local data at a client device and the global version of the machine learning model. A secure center different from the local center is generated based, at least in part, on information about secure centers shared by a plurality of other devices participating in a federated learning scheme. Information about the local version of the machine learning model and information about the secure center is transmitted by the local device to the server.

Abstract: A processor-implemented method for compressing a deep neural network model includes receiving an initial neural network model. The initial neural network is pruned based on a first threshold to generate a pruned network and a set of pruned weights. A quantization process is applied to the pruned network to produce a pruned and quantized network. A teacher model is generated by incorporating the pruned set of weights with the pruned network. In addition, an initial student model is generated from the quantized and pruned network. The initial student model is trained using the teacher model to output a trained student model.

Abstract: A method for generating a personalized model includes receiving one or more personal data samples from a user. A prototype of a personal identity is generated based on the personal data samples. The prototype of the personal identity is trained to reflect personal characteristics of the user. A network graph is generated based on the prototype of the personal identity. One or more channels of a global network are pruned based on the network graph to produce the personalized model.

Abstract: A computer-implemented method of operating an artificial neural network for processing data having a frequency dimension includes receiving an input. The audio input may be separated into one or more subgroups along the frequency dimension. A normalization may be performed on each subgroup. The normalization for a first subgroup the normalization is performed independently of the normalization a second subgroups. An output such as a keyword detection indication, is generated based on the normalized subgroups.