公开(公告)号：WO2023062461A1

公开(公告)日：2023-04-20

申请号：PCT/IB2022/058950

申请日：2022-09-22

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

Inventor： SUNEJA, Sahil ,  ZHUANG, Yufan ,  ZHENG, Yunhui ,  MORARI, Alessandro ,  LAREDO, Jim

IPC: G06N3/08 ,  G06N3/00 ,  G06N5/02 ,  G06N5/00

Abstract: Techniques regarding augmenting one or more training datasets for training one or more AI models are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise training augmentation component that can generate an augmented training dataset for training an artificial intelligence model by extracting a simplified source code sample from a source code sample comprised within a training dataset.

公开(公告)号：WO2023047197A1

公开(公告)日：2023-03-30

申请号：PCT/IB2022/056099

申请日：2022-06-30

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

Inventor： RAHIMI, Abbas ,  SEBASTIAN, Abu

IPC: G06F15/16

Abstract: The invention is notably directed to a sensor system (100) for performing distributed sensing and classification of sensor data. The sensor system (100) comprises a set of distributed sensor nodes (111) for sensing the sensor data. The sensor system (100) is configured to encode the sensor data of each sensor node of a set of distributed sensor nodes (111) for sensing the sensor data as high-dimensional vectors (V1,V2,......VS) and to transmit the high-dimensional vectors (V1,V2,......VS) over a respective link between the respective sensor node and a receiver system (130). The sensor system (100) is further configured to superpose the high-dimensional vectors (V1,V2,......VS) of the sensor data from the set of sensor nodes (111) by physical superposition, thereby generating a superposed high-dimensional vector and to classify the superposed high-dimensional vectors (V1,V2,......VS) at the receiver system (130).

公开(公告)号：WO2022130088A1

公开(公告)日：2022-06-23

申请号：PCT/IB2021/061161

申请日：2021-12-01

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

Inventor： GIBLIN, Christopher ,  ROONEY, John ,  FROESE, Florian ,  VETSCH, Pascal

IPC: H04L9/16

Abstract: A method for a cryptographic key rotation in a publish-subscribe system providing a broker (202) service for routing stored encrypted messages (204) to one or more subscribers of the topic (201) to which the routed messages(204) are assigned. The routing comprises decrypting the stored encrypted messages (204). The cryptographic key rotation comprises a re-encryption of the stored messages (204) using a cryptographic replacement key. The re-encryption is executed by an encryption module of the publish- subscribe system as a background process, while the broker (202) service is continued.

公开(公告)号：WO2022123370A1

公开(公告)日：2022-06-16

申请号：PCT/IB2021/060784

申请日：2021-11-21

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION [US]/[US] ,  IBM ISRAEL SCIENCE AND TECHNOLOGY LTD. [IL]/[IL] (MG) ,  IBM (CHINA) INVESTMENT COMPANY LTD. [CN]/[CN] (MG)

Inventor： ROONEY, John ,  GARCÉS ERICE, Luis

IPC: G06F16/21

Abstract: An approach to finding data locations may be provided. A first synopses, which corresponds to first tabular data (14) may be provided. An auxiliary data storage system (2) may be scanned to identify second tabular data (24) stored in the auxiliay data storage system (2). A second synopses may be obtained, in which the synopses correspond to the second columns of second tabular data (24). The synopsis may be computed for each second column of the second columns, according to a numeric representation of contents of cells of said each second column. The computed synopsis may include a vector of m descriptors. The two sets of one or more descriptors may be compared. A subset of the descriptors of the second synopses may be compared with corresponding descriptors of the first synopses, to identify potential matches between the second tabular data (24) and the corpus of first tabular data (14).

公开(公告)号：WO2022069991A1

公开(公告)日：2022-04-07

申请号：PCT/IB2021/058530

申请日：2021-09-20

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

Inventor： ZLOTNICK, Aviad

IPC: G06N3/08

Abstract: Outlier detection using a Deep Neural Network (DNN) includes running a trained DNN model on an received input item. A first feature vector is extracted from the input item and quantized to discrete values. A first number of special t-way feature combinations are computed in the input item and compared against a computed threshold. Based on the comparison, the input item is flagged as an outlier and an alert is generated notifying of the flagged input item.

公开(公告)号：WO2022043778A1

公开(公告)日：2022-03-03

申请号：PCT/IB2021/056022

申请日：2021-07-06

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

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

IPC: H04L9/06 ,  H04L9/32

Abstract: In some embodiments, a processor may store a value and a hash of the value in a ledger associated with a blockchain network. The processor may construct a block hash from the hash of the value. The processor may validate a transaction by identifying that the hash of the value matches the hash of the value from a previous transaction. The processor may maintain integrity of the block hash. In some embodiments, a processor may provide a redact transaction for a blockchain network. The processor may initialize the redact transaction. The redact transaction may be associated with a validated transaction. The processor may identify a hash value indicated within the redact transaction and associated with the validated transaction. The processor may redact a value associated with the hash value.

公开(公告)号：WO2022029516A1

公开(公告)日：2022-02-10

申请号：PCT/IB2021/055247

申请日：2021-06-15

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

Inventor： ZORIN, Anton ,  LABBI, Abderrahim

IPC: G06F16/27 ,  G06F16/2365 ,  G06F16/24573 ,  G06F16/254 ,  G06F16/288 ,  G06F16/9024 ,  G06N5/022 ,  G06N5/025

Abstract: The exemplary embodiments disclose a system and method, a computer program product, and a computer system. The exemplary embodiments may include receiving a data analysis request, using a knowledge graph for determining a source dataset based on the received data analysis request, wherein the knowledge graph represents an extract, transform and load (ETL) based ontology, wherein the knowledge graph comprises nodes representing entities and edges representing relationships between the entities, and wherein the entities are instances of concepts of the ETL based ontology, building an ETL workflow for processing the source dataset in accordance with a data lineage associated with the source dataset in the knowledge graph, and executing the ETL workflow.

公开(公告)号：WO2022003454A1

公开(公告)日：2022-01-06

申请号：PCT/IB2021/055058

申请日：2021-06-09

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

Inventor： MAPIYE, Darlington Shingirirai ,  MASHIYANE, James Junior ,  MULLER, Stephanie ,  MOKOATLE, Mpho ,  DLAMINI, Gciniwe

IPC: C12Q1/68 ,  G06N3/0454 ,  G06N3/088 ,  G16B30/10 ,  G16B40/30

Abstract: A method for automatically downsampling DNA sequence data using variational autoencoders and preserving genomic integrity of an original file is disclosed. The method comprises: executing, by an encoder(142), bootstrapping on genomic sequence data to produce resamples; assessing, by the encoder(142), unrepresentativeness and self-inconsistency of the resamples and selecting a representative resample according to the assessment; building, by a modified encoder(142), vector representations from genotype likelihoods based on the selected representative sample; integrating, by an analytics engine, mapping positional information and the genotype likelihoods to identify an optimum vector representation of a resample; and decoding, by a modified decoder(144), the identified optimum vector representation of the resample to obtain a down-sampled read file that resembles and maintains the genornic integrity of the original file.

公开(公告)号：WO2021229312A1

公开(公告)日：2021-11-18

申请号：PCT/IB2021/052265

申请日：2021-03-18

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

Inventor： LUUS, Francois ,  RIEGEL, Ryan ,  AKHALWAYA, Ismail Yunus ,  KHAN, Naweed Aghmad ,  VOS, Etienne ,  MAKONDO, Ndivhuwo

IPC: G06N3/04 ,  G06N3/0481 ,  G06N3/08

Abstract: Maximum expressivity can be received representing a ratio between maximum and minimum input weights to a neuron of a neural network implementing a weighted real-valued logic gate. Operator arity can be received associated with the neuron. Logical constraints associated with the weighted real-valued logic gate can be determined in terms of weights associated with inputs to the neuron, a threshold-of-truth, and a neuron threshold for activation. The threshold-of-truth can be determined as a parameter used in an activation function of the neuron, based on solving an activation optimization formulated based on the logical constraints, the activation optimization maximizing a product of expressivity representing a distribution width of input weights to the neuron and gradient quality for the neuron given the operator arity and the maximum expressivity. The neural network of logical neurons can be trained using the activation function at the neuron, the activation function using the determined threshold-of-truth.

公开(公告)号：WO2023089387A1

公开(公告)日：2023-05-25

申请号：PCT/IB2022/058690

申请日：2022-09-15

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

Inventor： PFEFFERLE, Jonas ,  IOANNOU, Nikolas ,  CASTANOS, Jose Gabriel ,  HARNIK, Danny ,  SHAH, Gautam

IPC: G06F9/50 ,  G06F9/455

Abstract: Live migration of a virtual machine (VM) includes establishing multipath connections between the VM and functions of host interface on a source host. The multipath connections include a passthrough path and a software-virtualized (or emulated) path provided by a hypervisor of the source host. A failover of the passthrough path to the emulated path is executed, and a state of the emulated path is thereafter saved. On a host interface of a destination host, functions corresponding to those of the source host are exposed. The VM is then migrated from the source host to the destination host. The VM resumes host interface communication with the host interface of the destination host from the saved state via an emulated path provided by a hypervisor of the destination host. After resuming communication, a passthrough path of communication between the VM and the host interface of the destination host is established.