公开(公告)号：WO2021094853A1

公开(公告)日：2021-05-20

申请号：PCT/IB2020/059643

申请日：2020-10-14

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： WANG, Weichen ,  SALKELD, Eliza ,  LU, Lucy ,  HAYES, Shanna

IPC: G06F9/44

Abstract: A detection modeling system has a processing device and a memory coupled to the processing device. The detection modeling system is configured to select a detection model for assessing model degradation, the detection model being an analytical model for analyzing data and identifying target events, perform a survival analysis on a plurality of like detection models, determine a survival metric for the selected detection model based on the survival analysis, and compare the survival metric to a threshold value and alerting to model degradation based on the comparison.

公开(公告)号：WO2021090100A1

公开(公告)日：2021-05-14

申请号：PCT/IB2020/059927

申请日：2020-10-22

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM (CHINA) INVESTMENT COMPANY LTD. ,  IBM ISRAEL - SCIENCE & TECHNOLOGY LTD.

Inventor： MANEVICH, Yacov ,  BARGER, Artem ,  MEIR, Hagar

IPC: G06F21/64 ,  H04L29/06 ,  H04L9/06

Abstract: An example operation may include one or more of retrieving a block hash of a data block stored on a blockchain, randomly determining a subset of peer organizations from a blockchain network of the blockchain to be endorsers based on a value of the block hash, transmitting a blockchain storage request from a client to the randomly determined subset of endorser peer organizations, and collecting simulated responses from the randomly determined subset of endorser peer organizations into a storage request proposal.

公开(公告)号：WO2021090091A1

公开(公告)日：2021-05-14

申请号：PCT/IB2020/059698

申请日：2020-10-15

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： CORREIA VILLA REAL, Lucas ,  SANTOS, Marcelo Nery dos ,  SOUZA, Renan Francisco Santos

IPC: G06F3/06

Abstract: Techniques for automatically recreating data removed from storage are disclosed. A request to access data at a first storage location is received. It is determined, based on a catalog and using a computer processor, that the data has been removed from the first storage location. In response the data at the first storage location is automatically recreated using the computer processor. The recreating includes identifying a second storage location for the data, based on the catalog, retrieving the data from the second storage location, and storing the data at the first storage location. The catalog is updated to reflect recreating the data at the first storage location.

公开(公告)号：WO2021084434A1

公开(公告)日：2021-05-06

申请号：PCT/IB2020/060090

申请日：2020-10-28

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： LLAMAS VIRGEN, Paul ,  GARCIA DELGADO, Carolina ,  OROZCO CERVANTES, Humberto

IPC: H04L9/32 ,  G06F21/46

Abstract: Described are techniques for collaboration-based authentication including a method comprising storing a user profile comprising information related to a user account, a plurality of user devices, and a user calendar. The method further comprises detecting an authentication attempt associated with the user account and retrieving device information associated with a first device initiating the authentication attempt, a first location of the first device, a second location of a second device of the plurality of user devices, and a scheduled location based on the user calendar. The method further comprises determining that the authentication attempt comprises a security risk based on the device information associated with the first device, the first location of the first device, the second location of the second device, and the scheduled location. The method further comprises performing a mitigation action in response to determining that the authentication attempt comprises the security risk.

公开(公告)号：WO2021084366A1

公开(公告)日：2021-05-06

申请号：PCT/IB2020/059751

申请日：2020-10-16

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： SUN, Jonathan Zanhong

IPC: H01L43/08

Abstract: A modified double magnet tunnel junction (mDMTJ) structure (100) is provided which includes a non-magnetic, spin-conducting metallic layer (106) sandwiched between a magnetic free layer (108) and a first tunnel barrier layer (104); the first tunnel barrier layer (104) contacts a first magnetic reference layer (102). A second tunnel barrier layer (110) is located on the magnetic free layer (108) and a second magnetic reference layer (112) is located on the second tunnel barrier layer (110). The mDMTJ structure (100) exhibits efficient switching and speedy readout.

公开(公告)号：WO2021079211A1

公开(公告)日：2021-04-29

申请号：PCT/IB2020/058933

申请日：2020-09-24

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： DUTTA, Ashim ,  ARNOLD, John ,  METZLER, Dominik

IPC: H01L21/768 ,  H01L23/522

Abstract: Method of forming self-aligned vias (2302, 2402) at line ends includes: patterning even/odd metal lines (202, 302) including using a first/second hardmask (102, 104); cutting the hardmask (102, 104) and a select metal line (202, 302), even or odd, using a cut mask (402, 1102) having a window (404) that exposes the hardmask (102, 104) over a cut region of the select metal line (202, 302); enlarging the window (404) to expose the hardmask (102, 104) on either side of the out region; selectively etching the hardmask (102, 104) using the enlarged window (1104) to form a T-shaped cavity (902, 1502) within the out region; filling the T-shaped cavity (902, 1502) with a gap fill dielectric (2002); removing the hardmask (102, 104); and recessing the metal lines (202, 302), wherein the gap fill dielectric (2002) overhangs portions of the select metal line (202, 302) that, by the recessing, form the self-aligned vias (2302, 2402) at ends of the metal lines (202, 302). A structure is also provided.

公开(公告)号：WO2021079210A1

公开(公告)日：2021-04-29

申请号：PCT/IB2020/058917

申请日：2020-09-24

Applicant: INTERNATIONAL BUSINESS MACHINES CORPOFIATION ,  IBM ISRAEL - SCIENCE & TECHNOLOGY LTD ,  IBM (CHINA) INVESTMENT COMPANY LTD

Inventor： ALBRECHT, Conrad ,  HAMANN, Hendrik ,  KLEIN, Levente ,  LU, Siyuan ,  SHARATHCHANDRA, Pankanti ,  ZHOU, Wang

IPC: G06T7/00 ,  G06K9/62

Abstract: Methods and systems for managing vegetation include training a machine learning model based on an image of a training data region before a weather event, an image of the training data region after the weather event, and information regarding the weather event. A risk score is generated for a second region using the trained machine learning model based on an image of the second region and predicted weather information for the second region. The risk score is determined to indicate high-risk vegetation in the second region. A corrective action is performed to reduce the risk of vegetation in the second region.

公开(公告)号：WO2021074717A1

公开(公告)日：2021-04-22

申请号：PCT/IB2020/058832

申请日：2020-09-22

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： JANTA-POLCZYNSKI, Barnim Alexander

IPC: G02B6/40 ,  G02B6/38

Abstract: Disclosed are an interlacing boot (102) and methods of using the same to automatically interleave optical fibers in a two-row array, such from a two rows ferrule. The optical fibers (202) are inserted into a first end of an interlacing boot (102) in a first direction. The interlacing boot (102) can include a guiding structure (206) having one or more channels (204). Each channel (204) can be adapted to receive a single optical fiber. Each channel (204) can include a first and and a second end, and the second end can be offset with respect to the first end in a second direction orthogonal to the first direction. The interlacing boot (102) can be pushed towards the ferrule (100) to feed the optical fibers (202) through the guiding structure (206). The first row of fibers can be physically offset from and interlaced with the second row of fibers by the guiding structure.

公开(公告)号：WO2021070135A1

公开(公告)日：2021-04-15

申请号：PCT/IB2020/059511

申请日：2020-10-09

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： BELLER, Charles ,  JOSHI, Akanksha ,  SIVASWAMY, Hemant, Kumar ,  NAMBIAR, Sunil

IPC: G10L15/26

Abstract: A method modifies and utilizes a text document based on an audio file that has captured audio proceedings relevant to the text document. One or more processors and/or a user segment the text document into topic-based text segments. Processor(s) receive an audio file that is relevant to the text document, convert the audio file into text, and match a specific topic-based text segment from the topic-based text segments to a specific topic-based audio segment from the topic-based audio segments. In response to identifying a difference between content in the specific topic-based text segment and content in the specific topic-based audio segment, processor(s) and/or the user update the specific topic-based text segment with the content in the specific topic-based audio segment to create an updated version of the text document.

公开(公告)号：WO2021070108A1

公开(公告)日：2021-04-15

申请号：PCT/IB2020/059467

申请日：2020-10-08

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： BRAMAN, Nathaniel, Mason ,  MARVAST, Ehsan, Dehghan ,  BEYMER, David, James

IPC: G16H50/20 ,  G06N3/02

Abstract: Systems and methods for developing a disease detection model. One method includes training the model using an image study and an associated disease label mined from a radiology report. The image study including a sequence of a plurality of two-dimensional slices of a three-dimensional image volume, and the model including a convolutional neural network layer (104) and a convolutional long short-term memory layer (108). Training the model includes individually extracting a set of features from each of the plurality of two-dimensional slices using the convolutional neural network layer (104), sequentially processing the features extracted by the convolutional neural network layer (104) for each of the plurality of two-dimensional slices using the convolutional long short-term memory layer (108), processing output from the convolutional long short-term memory layer (108) for a sequentially last of the plurality of two-dimensional slices to generate a probability of the disease, and updating the model based on comparing the probability to the label.