公开(公告)号：US12175359B2

公开(公告)日：2024-12-24

申请号：US16558585

申请日：2019-09-03

Applicant: International Business Machines Corporation

Inventor： Xiao Sun ,  Jungwook Choi ,  Naigang Wang ,  Chia-Yu Chen ,  Kailash Gopalakrishnan

IPC: G06N3/08 ,  G06N3/04

Abstract: An apparatus for training and inferencing a neural network includes circuitry that is configured to generate a first weight having a first format including a first number of bits based at least in part on a second weight having a second format including a second number of bits and a residual having a third format including a third number of bits. The second number of bits and the third number of bits are each less than the first number of bits. The circuitry is further configured to update the second weight based at least in part on the first weight and to update the residual based at least in part on the updated second weight and the first weight. The circuitry is further configured to update the first weight based at least in part on the updated second weight and the updated residual.

公开(公告)号：US11977974B2

公开(公告)日：2024-05-07

申请号：US15827465

申请日：2017-11-30

Applicant: International Business Machines Corporation

Inventor： Chia-Yu Chen ,  Jungwook Choi ,  Kailash Gopalakrishnan ,  Suyog Gupta ,  Pritish Narayanan

IPC: G06N3/08 ,  G06F17/14

CPC classification number: G06N3/08 ,  G06F17/147

Abstract: A system, having a memory that stores computer executable components, and a processor that executes the computer executable components, reduces data size in connection with training a neural network by exploiting spatial locality to weight matrices and effecting frequency transformation and compression. A receiving component receives neural network data in the form of a compressed frequency-domain weight matrix. A segmentation component segments the initial weight matrix into original sub-components, wherein respective original sub-components have spatial weights. A sampling component applies a generalized weight distribution to the respective original sub-components to generate respective normalized sub-components. A transform component applies a transform to the respective normalized sub-components. A cropping component crops high-frequency weights of the respective transformed normalized sub-components to yield a set of low-frequency normalized sub-components to generate a compressed representation of the original sub-components.

公开(公告)号：US11853713B2

公开(公告)日：2023-12-26

申请号：US15954891

申请日：2018-04-17

Applicant: International Business Machines Corporation

Inventor： Pin-Yu Chen ,  Lingfei Wu ,  Chia-Yu Chen ,  Yada Zhu

IPC: G06F16/00 ,  G06F7/20 ,  G06N20/00 ,  G06F16/23 ,  G06F16/901

CPC classification number: G06F7/20 ,  G06F16/2379 ,  G06F16/9024 ,  G06N20/00

Abstract: Techniques that facilitate graph similarity analytics are provided. In one example, a system includes an information component and a similarity component. The information component generates a first information index indicative of a first entropy measure for a first graph-structured dataset associated with a machine learning system. The information component also generates a second information index indicative of a second entropy measure for a second graph-structured dataset associated with the machine learning system. The similarity component determines similarity between the first graph-structured dataset and the second graph-structured dataset based on a graph similarity computation associated with the first information index and the second information index.

公开(公告)号：US11816549B2

公开(公告)日：2023-11-14

申请号：US16204770

申请日：2018-11-29

Applicant: International Business Machines Corporation

Inventor： Wei Zhang ,  Chia-Yu Chen

IPC: G06N3/098 ,  G06N20/20 ,  G06N7/08

CPC classification number: G06N20/20 ,  G06N3/098 ,  G06N7/08

Abstract: Systems, computer-implemented methods, and computer program products to facilitate gradient weight compression are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a pointer component that can identify one or more compressed gradient weights not present in a first concatenated compressed gradient weight. The computer executable components can further comprise a compression component that can compute a second concatenated compressed gradient weight based on the one or more compressed gradient weights to update a weight of a learning entity of a machine learning system.

公开(公告)号：US11797851B2

公开(公告)日：2023-10-24

申请号：US18068637

申请日：2022-12-20

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Chia-Yu Chen ,  Jui-Hsin Lai ,  Ko-Tao Lee ,  Li-Wen Hung

IPC: G11C16/04 ,  G06N3/08 ,  G06N5/04 ,  G11C11/419 ,  G11C11/412

CPC classification number: G06N3/08 ,  G06N5/04 ,  G11C11/412 ,  G11C11/419

Abstract: A Static Random Access Memory (SRAM) device in a binary neural network is provided. The SRAM device includes an SRAM inference engine having an SRAM computation architecture with a forward path that include multiple SRAM cells forming a chain of SRAM cells such that an output of a given one of the multiple SRAM cells is an input to a following one of the multiple SRAM cells. The SRAM computation architecture is configured to compute a prediction from an input. The SRAM computation architecture is configured to store ternary data and perform local computations on the ternary data.

公开(公告)号：US11599785B2

公开(公告)日：2023-03-07

申请号：US16188922

申请日：2018-11-13

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Chia-Yu Chen ,  Jui-Hsin Lai ,  Ko-Tao Lee ,  Li-Wen Hung

IPC: G11C16/04 ,  G06N3/08 ,  G06N5/04 ,  G11C11/419 ,  G11C11/412

Abstract: A Static Random Access Memory (SRAM) device in a binary neural network is provided. The SRAM device includes an SRAM inference engine having an SRAM computation architecture with a forward path that include multiple SRAM cells. The multiple SRAM cells are configured to form a chain of SRAM cells such that an output of a given one of the multiple SRAM cells is an input to a following one of the multiple SRAM cells. The SRAM computation architecture is configured to compute a prediction from an input.

公开(公告)号：US11586912B2

公开(公告)日：2023-02-21

申请号：US16657263

申请日：2019-10-18

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Chia-Yu Chen ,  Pin-Yu Chen ,  Mingu Kang ,  Jintao Zhang

IPC: G06N3/08 ,  G06N3/063

Abstract: Methods, systems, and circuits for training a neural network include applying noise to a set of training data across wordlines using a respective noise switch on each wordline. A neural network is trained using the noise-applied training data to generate a classifier that is robust against adversarial training.

公开(公告)号：US20220092407A1

公开(公告)日：2022-03-24

申请号：US17029506

申请日：2020-09-23

Applicant: International Business Machines Corporation ,  National Tsing Hua University

Inventor： Pin-Yu Chen ,  Sijia Liu ,  Chia-Yu Chen ,  I-Hsin Chung ,  Tsung-Yi Ho ,  Yun-Yun Tsai

IPC: G06N3/08 ,  G06N20/00 ,  G06K9/62

Abstract: Transfer learning in machine learning can include receiving a machine learning model. Target domain training data for reprogramming the machine learning model using transfer learning can be received. The target domain training data can be transformed by performing a transformation function on the target domain training data. Output labels of the machine learning model can be mapped to target labels associated with the target domain training data. The transformation function can be trained by optimizing a parameter of the transformation function. The machine learning model can be reprogrammed based on input data transformed by the transformation function and a mapping of the output labels to target labels.

公开(公告)号：US10552486B2

公开(公告)日：2020-02-04

申请号：US15165144

申请日：2016-05-26

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Chia-Yu Chen ,  Pei-Yun Hsueh ,  Jui-Hsin Lai ,  Yinglong Xia

IPC: G06F16/901

Abstract: A computer-implemented method, computer program product, and system for determination of critical parts and component correlations in a circuit using a correlation graph and centrality analysis including; receiving a circuit layout portion of a larger circuit layout, converting the circuit layout portion into a correlation graph representing components as nodes and connecting wires as edges, determining, using ground truth and Naïve Bayes to determine correlation weighting, scaling the correlation graph to represent the larger circuit, and presenting the larger correlation graph on a graphical user interface (GUI).

公开(公告)号：US10001516B2

公开(公告)日：2018-06-19

申请号：US15014350

申请日：2016-02-03

Applicant: International Business Machines Corporation

Inventor： Chia-Yu Chen ,  Shu-Jen Han

IPC: G01R27/08 ,  G01R27/02 ,  G01N27/414 ,  G01N27/02 ,  G01N33/00

CPC classification number: G01R27/02 ,  G01N27/02 ,  G01N27/4148 ,  G01N33/0031

Abstract: Frequency division multiplexing-based techniques for FET-based sensor arrays are provided. In one aspect, a sensor device includes: an array of FET-based sensors, wherein the sensors are grouped into multiple channels, and wherein each of the sensors includes an insulator on a substrate, a local gate embedded in the insulator, a channel material over the local embedded gate, and source and drain electrodes in contact with opposite ends of the channel material, and wherein a surface of the channel material is functionalized to react with at least one target molecule. The sensors in a given channel can be modulated (via the local gate) to enable the signal read out from the channel to be divided in the frequency domain based on the different frequencies used to modulate the sensors.