公开(公告)号：US20210081760A1

公开(公告)日：2021-03-18

申请号：US16572752

申请日：2019-09-17

Applicant: International Business Machines Corporation

Inventor： Li Cao ,  Guang Cheng Li ,  Rong Yan ,  Qi Ming Teng ,  Yubo Li ,  Cheng Fang Wang

IPC: G06N3/04 ,  G06N3/10 ,  G06N3/08 ,  G06F11/34 ,  G06K9/62 ,  G06F17/15

Abstract: Disclosed are a computer-implemented method, a system, and a computer program product for system-level tunable parameter identification. Performance characteristic data for an application to be tuned can be obtained by one or more processing units. At least one system-level tunable parameter for the application to be tuned can be identified by one or more processing units based on the obtained performance characteristic data for the application to be tuned and a pattern between training performance characteristic data and a set of training system-level parameter-related correlation coefficients. The set of training system-level parameter-related correlation coefficients can be respective correlation coefficients of system-level tunable parameters with respect to at least one performance metric.

公开(公告)号：US11907678B2

公开(公告)日：2024-02-20

申请号：US17093879

申请日：2020-11-10

Applicant: International Business Machines Corporation

Inventor： Fan Wang ,  Li Cao ,  Rui Wang ,  Lei Gao

IPC: G06F40/45 ,  G06F40/51 ,  G10L15/26 ,  G06F40/30 ,  G06F40/284 ,  G06F40/58

CPC classification number: G06F40/58 ,  G06F40/284 ,  G06F40/30 ,  G06F40/45 ,  G06F40/51 ,  G10L15/26

Abstract: A machine translation system, a ChatOps system, a method for a context-aware language machine identification, and computer program product. One embodiment of the machine translation system may include a density calculator. The density calculator may be adapted to calculate a part of speech (POS) density for a plurality of word tokens in an input text, calculate a knowledge density for the plurality of word tokens, and calculate an information density for the plurality of word tokens using the POS density and the knowledge density. In some embodiments, the machine translation system may further comprise a sememe attacher and a context translator.

公开(公告)号：US11775317B2

公开(公告)日：2023-10-03

申请号：US17245042

申请日：2021-04-30

Applicant: International Business Machines Corporation

Inventor： Qin Yue Chen ,  Li Cao ,  Fei Fei Li ,  Han Su

IPC: G06F9/445 ,  G06F11/34 ,  G06F8/41 ,  G06N3/063

CPC classification number: G06F9/44505 ,  G06F8/443 ,  G06F8/445 ,  G06F11/3428 ,  G06N3/063 ,  G06F2201/81

Abstract: Embodiments for locating performance hot spots include collecting sample data having instruction addresses, the sample data being for a neural network model and determining instructions in the instruction addresses that are performance hot spots. A listing file is used to map the instructions of the sample data that are performance hot spots to locations in a lower-level intermediate representation. A mapping file is used to map the locations of the lower-level intermediate representation that are performance hot spots to operations in one or more higher-level representations, one or more of the operations corresponding to the performance hot spots, the mapping file being generated from compiling the neural network model.

公开(公告)号：US11683317B2

公开(公告)日：2023-06-20

申请号：US17032010

申请日：2020-09-25

Applicant: International Business Machines Corporation

Inventor： Li Cao ,  Ze Ming Zhao ,  Qing Li ,  Yi Shan Jiang ,  Cheng Fang Wang

IPC: H04L9/40 ,  G06N20/00 ,  G06Q20/32 ,  G06Q40/02 ,  G06Q20/10 ,  G06Q20/12 ,  G06Q20/40

CPC classification number: H04L63/102 ,  G06N20/00 ,  H04L63/104 ,  H04L63/108 ,  G06Q20/108 ,  G06Q20/127 ,  G06Q20/3223 ,  G06Q20/4037 ,  G06Q40/02

Abstract: Embodiments of the present invention relate to methods, systems, and computer program products for user behavior management. In embodiments, a group of states of a user of an application system within a previous time period may be obtained. A state in the group of states may be associated with a privilege of the user for accessing resources in the application system during the previous time period. A feature of the user may be generated based on the group of states. A privilege of the user at a current time may be managed in the application system based on the feature. With these embodiments, the user behavior may be managed according to various aspect of the user's historical states and thus the user may be managed in a more accurate and effective manner.

公开(公告)号：US11620205B2

公开(公告)日：2023-04-04

申请号：US17073504

申请日：2020-10-19

Applicant: International Business Machines Corporation

Inventor： Li Cao ,  Xing Xing Shen ,  Zhi Li ,  He Jiang Jia ,  Bo Tong Liu ,  Xiao Dong Li ,  Sheng Jie BJ Han

IPC: G06F11/34 ,  G06F17/15

Abstract: A computer-implemented method for determining influence of applications on system performance includes collecting, by a processor, for several applications that are executing on a computing system, respective instrumentation data during multiple time-segments. The method further includes determining, for each of the applications, a performance value and a robustness value for each of the time-segments based on the respective instrumentation data. Further, using the performance value and robustness value for each time-segment, multiple health-waveforms are generated, where a health-waveform is generated for each respective application. The method further includes determining, by the processor, an influence-factor of a first application on a second application, the first application and the second application are executing on the computing system. The method further includes adjusting, by the processor, allocation of a computer resource by releasing the computer resource from the first application and allocating the computer resource to the second application based on the influence-factor.

公开(公告)号：US20220383149A1

公开(公告)日：2022-12-01

申请号：US17330099

申请日：2021-05-25

Applicant: International Business Machines Corporation

Inventor： Li Cao ,  Yi Shan Jiang ,  Ze Ming Zhao ,  Hong Bo Peng

IPC: G06N5/04 ,  G06N20/00

Abstract: A computer-implemented method includes determining, by a master node, model update information at least based on a workload related to a task and a resource capacity of a computing environment. The model update information indicates respective model update suggestions for a plurality of inference models configured to perform the task. The method further includes distributing, by the master node, the model update information to a plurality of inference agents in the computing environment. The plurality of inference agents has a plurality of instances of the plurality of inference models executed thereon.

公开(公告)号：US20220350619A1

公开(公告)日：2022-11-03

申请号：US17245042

申请日：2021-04-30

Applicant: International Business Machines Corporation

Inventor： QIN YUE CHEN ,  Li Cao ,  Fei Fei Li ,  Han Su

IPC: G06F9/445 ,  G06F11/34 ,  G06F8/41 ,  G06N3/063

Abstract: Embodiments for locating performance hot spots include collecting sample data having instruction addresses, the sample data being for a neural network model and determining instructions in the instruction addresses that are performance hot spots. A listing file is used to map the instructions of the sample data that are performance hot spots to locations in a lower-level intermediate representation. A mapping file is used to map the locations of the lower-level intermediate representation that are performance hot spots to operations in one or more higher-level representations, one or more of the operations corresponding to the performance hot spots, the mapping file being generated from compiling the neural network model.

公开(公告)号：US20220292390A1

公开(公告)日：2022-09-15

申请号：US17197361

申请日：2021-03-10

Applicant: International Business Machines Corporation

Inventor： Li Cao ,  WeiFeng Zhang ,  Fei Fei Li ,  Ren Jie Feng ,  Han Su ,  Zhan Peng Huo ,  Zhong Hao Wang

IPC: G06N20/00 ,  G06N5/04

Abstract: Aspects of the invention include converting an artificial intelligence (AI) model generated in a first framework to a uniform exchange formatted model by engaging a master table to retrieve instructions for converting from the AI model to the uniform exchange formatted model in accordance with the first framework. The uniform exchange formatted model in compiled by engaging the master table to retrieve instructions for compiling the uniform exchange formatted model in accordance with the first framework. Data is received as an input to the compiled uniform exchange formatted model and an output is generated by engaging the master table to retrieve instructions for generating the output in accordance with the first framework.

公开(公告)号：US20220215328A1

公开(公告)日：2022-07-07

申请号：US17143524

申请日：2021-01-07

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Yang Chen ,  Li Cao ,  Ya Ju Yan ,  Zhou Kun ,  Li Hong Qi ,  Xiao Juan Chen

IPC: G06Q10/06 ,  G06N20/00

Abstract: An automated method for determining a complexity of a task. The method includes extracting data from the plurality of historical support tickets to generate training data. The method trains a complexity model to predict a complexity value of a task associated with a support ticket using the training data. The method predicts, using the complexity model, the complexity value of a new task associated with a new support ticket.

公开(公告)号：US20220188379A1

公开(公告)日：2022-06-16

申请号：US17117237

申请日：2020-12-10

Applicant: International Business Machines Corporation

Inventor： Jing Yan Ma ,  Chu Yun Tong ,  Li Cao ,  Peng Hui Jiang

IPC: G06F17/15 ,  G06F17/16 ,  G06F16/14

Abstract: Synching multiple streams in a complex enterprise product by collecting and analyzing stream dependency data. Collection and analysis of data for large scale and complex enterprise results in a multi-dimensional relationship diagram that highlights the interconnected dependencies of the streams. This allows enterprise software users to more easily determine and select which stream (or streams) will help the user to perform a given task.