专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US11169976B2 Data quality control 有权
公开(公告)号：US11169976B2
公开(公告)日：2021-11-09
申请号：US16438421
申请日：2019-06-11
申请人： International Business Machines Corporation
发明人： Sai Koti Reddy Danda , Pankaj S. Dayama , Priyanka Agrawal , Srikanth Govindaraj Tamilselvam , Amrita Saha
IPC分类号： G06F16/215 , G06F16/2458 , G06F16/27
摘要： An example operation may include one or more of receiving, by a monitoring peer, data collection requirements from a model builder node, querying, by the monitoring peer, a blockchain of a plurality of blockchains for data sets that match the data collection requirements, providing, by the monitoring peer, the data sets to the model builder node to be tested, receiving, by the monitoring peer, a performance feedback on the data sets from the model builder node; and updating, by the monitoring peer, a smart contract associated with the data sets on the blockchain based on the performance feedback.

2. 发明申请

US20200327420A1 DEBUGGING DEEP NEURAL NETWORKS 审中-公开
公开(公告)号：US20200327420A1
公开(公告)日：2020-10-15
申请号：US16380437
申请日：2019-04-10
申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION
发明人： Rahul Aralikatte , Srikanth Govindaraj Tamilselvam , Shreya Khare , Naveen Panwar , Anush Sankaran , Senthil Kumar Kumarasamy Mani
IPC分类号： G06N3/10 , G06F11/36 , G06N3/08 , G06N3/04
摘要： A method, computer system, and a computer program product for debugging a deep neural network is provided. The present invention may include identifying, automatically, one or more debug layers associated with a deep learning (DL) model design/code, wherein the identified one or more debug layers include one or more errors, wherein a reverse operation is introduced for the identified one or more debug layers. The present invention may then include presenting, to a user, a debug output based on at least one break condition, wherein in response to determining the at least one break condition is satisfied, triggering the debug output to be presented to the user, wherein the presented debug output includes a fix for the identified one or more debug layers in the DL model design/code and at least one actionable insight.

3. 发明申请

US20200304435A1 Online Forum Automated Feedback Generator 审中-公开
公开(公告)号：US20200304435A1
公开(公告)日：2020-09-24
申请号：US16358741
申请日：2019-03-20
申请人： International Business Machines Corporation
发明人： Tanmayee Narendra , Tarun Tater , Srikanth Govindaraj Tamilselvam , Senthil Kumar Kumarasamy Mani
IPC分类号： H04L12/58 , G06F9/54 , G06F11/34
摘要： Embodiments describe an approach for automatically generating feedback for an online forum. Embodiments determine if a user is using a solution to a problem, wherein the solution is posted on an online forum, and responsive to determining the user is using the solution, capturing environment information associated with the user's computing device. Additionally, embodiments determine if the solution solved the problem, and responsive to determining the solution solved the problem, automatically generate feedback associated with the solution, wherein the feedback comprises the environment information and information detailing that the solution solved the problem.

4. 发明授权

US09995588B2 Quantifying anomalous behavior by identifying anomalies at subtrajectories 有权
公开(公告)号：US09995588B2
公开(公告)日：2018-06-12
申请号：US14747883
申请日：2015-06-23
申请人： International Business Machines Corporation
发明人： Prithu Banerjee , Biplav Srivastava , Srikanth Govindaraj Tamilselvam
IPC分类号： G01C21/00 , G01C21/34
CPC分类号： G01C21/32
摘要： Methods and arrangements for identifying at least one anomaly in a path taken by a plurality of objects. A plurality of trajectories are input, wherein each trajectory comprises a data set indicative of a path taken by a plurality of objects from a starting point to an ending point, wherein the starting point and ending point are substantially similar for each trajectory. A plurality of sub-trajectories within the input trajectories are identified. There are identified, within the plurality of sub-trajectories, a set of sub-trajectories that are anomalous when compared to other sub-trajectories within the plurality of sub-trajectories, wherein the anomalous sub-trajectories deviate from a predetermined standard. A maximal anomalous sub-trajectory is identified from among the identified set of anomalous sub-trajectories. Other variants and embodiments are broadly contemplated herein.

5. 发明申请

US20160377441A1 QUANTIFYING ANOMALOUS BEHAVIOR BY IDENTIFYING ANOMALIES AT SUBTRAJECTORIES 有权
标题翻译：通过识别异常的异常来量化异常行为
公开(公告)号：US20160377441A1
公开(公告)日：2016-12-29
申请号：US14747883
申请日：2015-06-23
申请人： International Business Machines Corporation
发明人： Prithu Banerjee , Biplav Srivastava , Srikanth Govindaraj Tamilselvam
IPC分类号： G01C21/34
CPC分类号： G01C21/32
摘要： Methods and arrangements for identifying at least one anomaly in a path taken by a plurality of objects. A plurality of trajectories are input, wherein each trajectory comprises a data set indicative of a path taken by a plurality of objects from a starting point to an ending point, wherein the starting point and ending point are substantially similar for each trajectory. A plurality of sub-trajectories within the input trajectories are identified. There are identified, within the plurality of sub-trajectories, a set of sub-trajectories that are anomalous when compared to other sub-trajectories within the plurality of sub-trajectories, wherein the anomalous sub-trajectories deviate from a predetermined standard. A maximal anomalous sub-trajectory is identified from among the identified set of anomalous sub-trajectories. Other variants and embodiments are broadly contemplated herein.
摘要翻译：用于识别由多个对象取走的路径中的至少一个异常的方法和装置。输入多个轨迹，其中每个轨迹包括指示多个物体从起点到终点的路径的数据集，其中起始点和终点与每个轨迹基本相似。识别输入轨迹内的多个子轨迹。在多个子轨迹之间，识别与多个子轨迹内的其他子轨迹相比异常的一组子轨迹，其中异常子轨迹偏离预定标准。从所识别的一组异常子轨迹中识别最大异常子轨迹。本文中广泛考虑了其他变型和实施例。

6. 发明授权

US11605006B2 Deep-learning model catalog creation 有权
公开(公告)号：US11605006B2
公开(公告)日：2023-03-14
申请号：US16404141
申请日：2019-05-06
申请人： International Business Machines Corporation
发明人： Shreya Khare , Srikanth Govindaraj Tamilselvam , Anush Sankaran , Naveen Panwar , Rahul Rajendra Aralikatte , Senthil Kumar Kumarasamy Mani
IPC分类号： G06N20/00 , G06N5/02 , G06F8/41 , G06F40/205 , G06N5/022
摘要： One embodiment provides a method, including: mining a plurality of deep-learning models from a plurality of input sources; extracting information from each of the deep-learning models, by parsing at least one of (i) code corresponding to the deep-learning model and (ii) text corresponding to the deep-learning model; identifying, for each of the deep-learning models, operators that perform operations within the deep-learning model; producing, for each of the deep-learning models and from (i) the extracted information and (ii) the identified operators, an ontology comprising terms and features of the deep-learning model, wherein the producing comprises populating a pre-defined ontology format with features of each deep-learning model; and generating a deep-learning model catalog comprising the plurality of deep-learning models, wherein the catalog comprises, for each of the deep-learning models, the ontology corresponding to the deep-learning model.

7. 发明申请

US20220188362A1 DECENTRALIZED ONLINE MULTI-AGENT VISUAL QUESTION ANSWERING 有权
公开(公告)号：US20220188362A1
公开(公告)日：2022-06-16
申请号：US17118211
申请日：2020-12-10
申请人： International Business Machines Corporation
发明人： Ghulam Ahmed Ansari , Sai Koti Reddy Danda , Amrita Saha , Srikanth Govindaraj Tamilselvam
IPC分类号： G06F16/9032 , G06N20/00 , G06N5/04
摘要： An approach to real-time multi-agent visual question answering in a decentralized environment. The approach can include initializing a plurality of agents based on data associated with an image and a question about the image. The approach can include identifying a first portion of the plurality of agents to construct a grid of agents based on similarity between the first portion of agents. The approach can include determining a second portion of agents of the first portion of agents with edges between agents allowing sharing of agent concepts and agent parameters. The approach can include generate training trajectories of the grid of agents based on executing a collaborative rollout. Additionally, the approach can include optimizing agent policies of the grid of agents.

8. 发明授权

US11068943B2 Generating collaborative orderings of information pertaining to products to present to target users 有权
公开(公告)号：US11068943B2
公开(公告)日：2021-07-20
申请号：US16167703
申请日：2018-10-23
申请人： International Business Machines Corporation
发明人： Ramasuri Narayanam , Srikanth Govindaraj Tamilselvam , Sameep Mehta , Gyana Ranjan Parija
IPC分类号： G06Q30/00 , G06Q30/02 , G06F9/54 , G10L15/02
摘要： Methods, systems, and computer program products for generating collaborative orderings of information pertaining to products to present to target users are provided herein. A computer-implemented method includes generating, for each of multiple target users, a personalized initial ordering of multiple statements pertaining to an entity by analyzing the statements and user-specific information; calculating, for each of the target users, a utility score that comprises (i) similarity between a consensus ordering of the statements and the given target user's personalized initial ordering of the statements, and (ii) similarity between the other target users' personalized initial orderings of the statements; generating a final ordering of the statements by enabling, based on the calculated utility scores, agents representing the target users to exchange revised orderings of the multiple statements via one or more automated communication iterations; and outputting the statements, in the final ordering, to the target users.

9. 发明授权

US10938752B2 Online forum automated feedback generator 有权
公开(公告)号：US10938752B2
公开(公告)日：2021-03-02
申请号：US16358741
申请日：2019-03-20
申请人： International Business Machines Corporation
发明人： Tanmayee Narendra , Tarun Tater , Srikanth Govindaraj Tamilselvam , Senthil Kumar Kumarasamy Mani
IPC分类号： G06F15/16 , H04L12/58 , G06F11/34 , G06F9/54
摘要： Embodiments describe an approach for automatically generating feedback for an online forum. Embodiments determine if a user is using a solution to a problem, wherein the solution is posted on an online forum, and responsive to determining the user is using the solution, capturing environment information associated with the user's computing device. Additionally, embodiments determine if the solution solved the problem, and responsive to determining the solution solved the problem, automatically generate feedback associated with the solution, wherein the feedback comprises the environment information and information detailing that the solution solved the problem.

10. 发明授权

US10885347B1 Out-of-context video detection 有权
公开(公告)号：US10885347B1
公开(公告)日：2021-01-05
申请号：US16574980
申请日：2019-09-18
申请人： International Business Machines Corporation
发明人： Tarun Tater , Anush Sankaran , Srikanth Govindaraj Tamilselvam , Naveen Panwar
IPC分类号： G06K9/00 , G06F16/48 , G06F16/483 , G06F16/783
摘要： One embodiment provides a method, comprising: identifying, using a processor, an individual in a video segment; ascertaining, from at least the video segment, a viewpoint expressed by the individual on a topic, wherein the viewpoint comprises at least one of: a stance of the individual and a sentiment of the individual toward the topic; identifying, using a processor, a superset video comprising the video segment, wherein the superset video is an originally published video; ascertaining, based at least in part on the superset video, an overarching viewpoint of the individual on the topic; determining whether an inconsistency exists between the viewpoint expressed by the individual in the video segment and the overarching viewpoint of the individual ascertained in the superset video; and alerting, responsive to determining that an inconsistency exists, a user that the video segment contains the inconsistency.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式