Amazon Technologies, Inc.

Adobe, Inc.

Karan Aggarwal

Anup Bandigadi Rao

Georgios Theocharous

Bharathan Balaji

Cameron Dean Whitehouse

Krzysztof Marcin Walczak

Sujan Kumar Gonugondla

Venkatesh Rao

Niranjini Rajagopal

Disclosed implementations determine transient distortions, amplitude distortions, and/or periodic distortions present in a voltage signal measured from an electrical circuit at a location and use those distortions to determine a device type of a device operating on the electrical circuit at the location. In some examples, the voltage signal may be filtered into a plurality of sub-bands and one or more two-dimensional representations generated based on transient distortions, amplitude distortions, and/or periodic distortions determined in each sub-band. The one or more 2D representations may be provided to a deep neural network to determine a device type that generated the distortions on the electrical circuit.

Device type detection

A single unified machine learning model (e.g., a neural network) is trained to perform both supervised event predictions and unsupervised time-varying clustering for a sequence of events (e.g., a sequence representing a user behavior) using sequences of events for multiple users using a combined loss function. The unified model can then be used for, given a sequence of events as input, predict a next event to occur after the last event in the sequence and generate a clustering result by performing a clustering operation on the sequence of events. As part of predicting the next event, the unified model is trained to predict an event type for the next event and a time of occurrence for the next event. In certain embodiments, the unified model is a neural network comprising a recurrent neural network (RNN) such as an Long Short Term Memory (LSTM) network.

Unified framework for dynamic clustering and discrete time event prediction

Growing community of inventors

Seattle, WA, United States of America

Karan Aggarwal