Robert Boschgmbh

Lg Electronics Inc.

Mohak Shah

Franz Laermer

Jiayi Liu

Unmesh Kurup

Thomas Rocznik

Seow Yuen Yee

Christian Peters

Jayanta Kumar Dutta

Jan Waldmann

Torsten Welfonder

Zubin Abraham

Seyed Hamid Mirebrahim

Mohammad Shokooi-Yekta

Samarth Tripathi

Fabian Henrici

A computer-implemented method includes: initializing model parameters for training a neural network; performing a forward pass and backpropagation for a first minibatch of training data; determining a new weight value for each of a plurality of nodes of the neural network using a gradient descent of the first minibatch; for each determined new weight value, determining whether to update a running mean corresponding to a weight of a particular node; based on a determination to update the running mean, calculating a new mean weight value for the particular node using the determined new weight value; updating the weight parameters for all nodes based on the calculated new mean weight values corresponding to each node; assigning the running mean as the weight for the particular node when training on the first minibatch is completed; and reinitializing running means for all nodes at a start of training a second minibatch.

Training a neural network using periodic sampling over model weights

A vital signs monitoring system includes a peak pattern detection module configured to output a peak prediction signal from sensor signals based on a peak prediction algorithm; a vital sign estimating module configured to estimate a vital sign based on the peak prediction signal; an activity and context detector module configured to output a context signal based on at least one environmental condition and/or activity level of the person; and a concept drift detection module configured to output a drift signal based on drift detected in the estimated vital sign. The peak pattern prediction module is configured to update the peak prediction algorithm based on the context signal and the drift signal.

Vital signs monitoring system and method

A method of operating a heating ventilation and air conditioning (HVAC) system of a structure, includes collecting first sensor data corresponding to a parameter of the HVAC system, collecting second sensor data that is different than the first sensor data, and generating clustered data by clustering the first sensor data and the second sensor data into a plurality of data clusters with a controller. The method also includes forming a transactional dataset based on at least the first sensor data, the second sensor data, and the clustered data with the controller, performing association rule mining (ARM) on the transactional dataset to generate a plurality of rules for each data cluster of the plurality of data clusters with the controller, and changing an operating characteristic of the HVAC system based on the plurality of rules with the controller to optimize the parameter.

System and method for optimizing energy use of a structure using a clustering-based rule-mining approach

A search framework for finding effective architectural building blocks for deep convolutional neural networks is disclosed. The search framework described herein utilizes a building block which incorporates branch and skip connections. At least some operations of the architecture of the building block are undefined and treated as hyperparameters which can be automatically selected and optimized for a particular task. The search framework uses random search over the reduced search space to generate a building block and repeats the building block multiple times to create a deep convolutional neural network.

Effective building block design for deep convolutional neural networks using search

Growing community of inventors

Dublin, CA, United States of America

Mohak Shah