-
公开(公告)号：US20200293882A1
公开(公告)日：2020-09-17
申请号：US16402204
申请日：2019-05-02
Applicant: Samsung Electronics Co., Ltd.
Inventor： Liu LIU , Georgios GEORGIADIS , Elham SAKHAEE , Weiran DENG
Abstract: A recurrent neural network that predicts blood glucose level includes a first long short-term memory (LSTM) network and a second LSTM network. The first LSTM network may include an input to receive near-infrared (NIR) radiation data and includes an output. The second LSTM network may include an input to receive the output of the first LSTM network and an output to output blood glucose level data based on the NIR radiation data input to the first LSTM network.
-
公开(公告)号：US20220129756A1
公开(公告)日：2022-04-28
申请号：US17572625
申请日：2022-01-10
Applicant: Samsung Electronics Co., Ltd.
Inventor： Weiran DENG , Georgios GEORGIADIS
Abstract: A technique to prune weights of a neural network using an analytic threshold function h(w) provides a neural network having weights that have been optimally pruned. The neural network includes a plurality of layers in which each layer includes a set of weights w associated with the layer that enhance a speed performance of the neural network, an accuracy of the neural network, or a combination thereof. Each set of weights is based on a cost function C that has been minimized by back-propagating an output of the neural network in response to input training data. The cost function C is also minimized based on a derivative of the cost function C with respect to a first parameter of the analytic threshold function h(w) and on a derivative of the cost function C with respect to a second parameter of the analytic threshold function h(w).
-
Patent Agency Ranking
公开(公告)号：US20210133278A1
公开(公告)日：2021-05-06
申请号：US16816247
申请日：2020-03-11
Applicant: Samsung Electronics Co., Ltd.
Inventor： Jun FANG , Joseph H. HASSOUN , Ali SHAFIEE ARDESTANI , Hamzah Ahmed Ali ABDELAZIZ , Georgios GEORGIADIS , Hui CHEN , David Philip Lloyd THORSLEY
Abstract: A method of quantizing an artificial neural network may include dividing a quantization range for a tensor of the artificial neural network into a first region and a second region, and quantizing values of the tensor in the first region separately from values of the tensor in the second region. Linear or nonlinear quantization may be applied to values of the tensor in the first region and the second region. The method may include locating a breakpoint between the first region and the second region by substantially minimizing an expected quantization error over at least a portion of the quantization range. The expected quantization error may be minimized by solving analytically and/or searching numerically.
-
公开(公告)号：US20200293893A1
公开(公告)日：2020-09-17
申请号：US16396619
申请日：2019-04-26
Applicant: Samsung Electronics Co., Ltd.
Inventor： Georgios GEORGIADIS , Weiran DENG
Abstract: A system and a method generate a neural network that includes at least one layer having weights and output feature maps that have been jointly pruned and quantized. The weights of the layer are pruned using an analytic threshold function. Each weight remaining after pruning is quantized based on a weighted average of a quantization and dequantization of the weight for all quantization levels to form quantized weights for the layer. Output feature maps of the layer are generated based on the quantized weights of the layer. Each output feature map of the layer is quantized based on a weighted average of a quantization and dequantization of the output feature map for all quantization levels. Parameters of the analytic threshold function, the weighted average of all quantization levels of the weights and the weighted average of each output feature map of the layer are updated using a cost function.
-
公开(公告)号：US20190050735A1
公开(公告)日：2019-02-14
申请号：US15724267
申请日：2017-10-03
Applicant: Samsung Electronics Co., Ltd.
Inventor： Zhengping JI , John Wakefield BROTHERS , Weiran DENG , Georgios GEORGIADIS
Abstract: A method is disclosed to reduce computational load of a deep neural network. A number of multiply-accumulate (MAC) operations is determined for each layer of the deep neural network. A pruning error allowance per weight is determined based on a computational load of each layer. For each layer of the deep neural network: a threshold estimator is initialized, and weights of each layer are pruned based on a standard deviation of all weights within the layer. A pruning error per weight is determined for the layer, and if the pruning error per weight exceeds a predetermined threshold, the threshold estimator is updated for the layer the weights of the layer are repruned using the updated threshold estimator and the pruning error per weight is re-determined until the pruning error per weight is less than the threshold. The deep neural network is then retrained.
-
公开(公告)号：US20230004813A1
公开(公告)日：2023-01-05
申请号：US17943176
申请日：2022-09-12
Applicant: Samsung Electronics Co., Ltd.
Inventor： Georgios GEORGIADIS , Weiran DENG
Abstract: A system and a method generate a neural network that includes at least one layer having weights and output feature maps that have been jointly pruned and quantized. The weights of the layer are pruned using an analytic threshold function. Each weight remaining after pruning is quantized based on a weighted average of a quantization and dequantization of the weight for all quantization levels to form quantized weights for the layer. Output feature maps of the layer are generated based on the quantized weights of the layer. Each output feature map of the layer is quantized based on a weighted average of a quantization and dequantization of the output feature map for all quantization levels. Parameters of the analytic threshold function, the weighted average of all quantization levels of the weights and the weighted average of each output feature map of the layer are updated using a cost function.
-
公开(公告)号：US20190180184A1
公开(公告)日：2019-06-13
申请号：US15894921
申请日：2018-02-12
Applicant: Samsung Electronics Co., Ltd.
Inventor： Weiran DENG , Georgios GEORGIADIS
Abstract: A technique to prune weights of a neural network using an analytic threshold function h(w) provides a neural network having weights that have been optimally pruned. The neural network includes a plurality of layers in which each layer includes a set of weights w associated with the layer that enhance a speed performance of the neural network, an accuracy of the neural network, or a combination thereof. Each set of weights is based on a cost function C that has been minimized by back-propagating an output of the neural network in response to input training data. The cost function C is also minimized based on a derivative of the cost function C with respect to a first parameter of the analytic threshold function h(w) and on a derivative of the cost function C with respect to a second parameter of the analytic threshold function h(w).
-
-
-
-
-
-
Search Results
7
records
(took 0.014 seconds)
