Abstract: An edge node included in a decentralized edge computing network generates a federated partial-data aggregation machine learning model. The edge node learns one or more model parameters via machine learning techniques and receives one or more auxiliary model parameters from additional edge nodes in the decentralized edge computing network, such as from a neighbor node group. In some cases, a neighbor node is identified in response to determining that the neighbor node includes a model with a relatively high estimated relevance to the model of the edge node. The edge node modifies the model to include an aggregation of the learned model parameters and the received auxiliary parameters. Respective weights are learned for the learned model parameters and also for the received auxiliary parameters. During training to learn the respective weights, the edge node stabilizes the learned model parameters and the received auxiliary parameters.

Abstract: A system includes a representation generator subsystem configured to execute a user representation model and a task prediction model to generate a user representation for a user. The user representation model receives user event sequence data comprises a sequence of user interactions with the system. The task prediction model is configured to train the user representation model. The user representation includes a vector of a predetermined size that represents the user event sequence data and is generated by applying the trained user representation model to the user event sequence data. A storage requirement of the user representation is less than a storage space requirement of the user event sequence data. The system includes a data store configured for storing the user representation in a user profile associated with the user.

Abstract: In implementations of systems for cloud-based resource allocation using meters, a computing device implements a resource system to receive resource data describing an amount of cloud-based resources reserved for consumption by client devices during a period of time and a total amount of cloud-based resources consumed by the client devices during the period of time. The resource system determines a consumption distribution using each meter included in a set of meters. Each of the consumption distributions allocates a portion of the total amount of the cloud-based resources consumed to each client device of the client devices. A particular meter used to determine a particular consumption distribution is selected based on a Kendall Tau coefficient of the particular consumption distribution. An amount of cloud-based resources to allocate for a future period of time is estimated using the particular meter and an approximate Shapley value.

Abstract: A system includes a representation generator subsystem configured to execute a user representation model and a task prediction model to generate a user representation for a user. The user representation model receives user event sequence data comprises a sequence of user interactions with the system. The task prediction model is configured to train the user representation model. The user representation includes a vector of a predetermined size that represents the user event sequence data and is generated by applying the trained user representation model to the user event sequence data. A storage requirement of the user representation is less than a storage space requirement of the user event sequence data. The system includes a data store configured for storing the user representation in a user profile associated with the user.

Abstract: Machine-learning model retargeting techniques are described. In one example, training data is generated by extrapolating feedback data collected from entities. These techniques supports an ability to identify a wider range of thresholds and corresponding entities than those available in the feedback data. This also provides an opportunity to explore additional thresholds than those used in the past through extrapolating operations outside of a range used to define a segment, for which, the feedback data is captured. These techniques also support retargeting of a machine-learning model for a secondary label that is different than a primary label used to initially train the machine-learning model.

Abstract: In implementations of systems for cloud-based resource allocation using meters, a computing device implements a resource system to receive resource data describing an amount of cloud-based resources reserved for consumption by client devices during a period of time and a total amount of cloud-based resources consumed by the client devices during the period of time. The resource system determines a consumption distribution using each meter included in a set of meters. Each of the consumption distributions allocates a portion of the total amount of the cloud-based resources consumed to each client device of the client devices. A particular meter used to determine a particular consumption distribution is selected based on a Kendall Tau coefficient of the particular consumption distribution. An amount of cloud-based resources to allocate for a future period of time is estimated using the particular meter and an approximate Shapley value.

Abstract: In some embodiments, a model training system trains a sample generation model configured to generate synthetic data entries for a dataset. The sample generation model includes a prior model for generating an estimated latent vector from a partially observed data entry, a proposal model for generating a latent vector from a data entry of the dataset and a mask corresponding to the partially observed data entry, and a generative model for generating the synthetic data entries from the latent vector and the partially observed data entry. The model training system trains the sample generation model to optimize an objective function that includes a first term determined using the synthetic data entries and a second term determined using the estimated latent vector and the latent vector. The trained sample generation model can be executed on a client computing device to service queries using the generated synthetic data entries.

Abstract: In some embodiments, a model training system trains a sample generation model configured to generate synthetic data entries for a dataset. The sample generation model includes a prior model for generating an estimated latent vector from a partially observed data entry, a proposal model for generating a latent vector from a data entry of the dataset and a mask corresponding to the partially observed data entry, and a generative model for generating the synthetic data entries from the latent vector and the partially observed data entry. The model training system trains the sample generation model to optimize an objective function that includes a first term determined using the synthetic data entries and a second term determined using the estimated latent vector and the latent vector. The trained sample generation model can be executed on a client computing device to service queries using the generated synthetic data entries.