International Business Machines Corporation

Toshiba Global Commerce Solutions Holdings Corporation

Other

Globalfoundries Inc.

The University of Queensland

Toshiba Global Commerce

Solutions Holdings Corporation

Sharathchandra UmapathiRao Pankanti

Rogerio Schmidt Feris

Arun Hampapur

Ankur Datta

Nalini Kanta Ratha

Quanfu Fan

Yun Zhai

Norman C Haas

Jonathan Hudson Connell, Ii

Ying Li

Russell Patrick Bobbitt

Charles A Otto

Lisa Maire Brown

Karthik Nandakumar

Sachiko Miyazawa

Andrew William Senior

Akira Yanagawa

Hoang Trinh

Myron Dale Flickner

Chung-Ching Lin

Erik Rueger

John Richard Smith

Frederik C Kjeldsen

Behjat Siddiquie

Gaurav Goswami

Hendrik F Hamann

Shai Halevi

Ying-Li Tian

Lisa Marie Brown

Conrad M Albrecht

Siyuan Lu

Rudolf Maarten Bolle

Aleksandr Y Aravkin

Zuoxuan Lu

Karthikeyan Natesan Ramamurthy

Xiaoyu Wang

Daniel Andre Vaquero

Yu Cheng

Lei Ding

Balamanohar Paluri

Roman Vaculin

Xuan Liu

Noel C Codella

Karthikeyan Shanmugam

Jonathan H Connell

Marcus Oliver Freitag

Kanthi K Sarpatwar

Tomas Krojzl

Frank D Fenhagen, Iv

Duan D Tran

Prasad Gabbur

Unsang Park

James R Culp

Arvind Kumar

Levente Klein

Rolf Schaefer

Robert Stephen Olyha, Jr

Wang Zhou

Janusz Marecki

Shelby Solomon Darnell

Ban Kawas

Yuichi Fujiki

Richa Singh

Mayank Vatsa

Jiyan Pan

Sayed Ali Emami

Mehrtash Harandi

Jingjing Liu

Itzhack Goldberg

Komminist S Weldemariam

Nadiya Kochura

Neil Sondhi

Kafai Lai

Thomas Guthrie Zimmerman

Huan Hu

Bruce Gordon Elmegreen

Chiao-Fe Shu

James Thomas Rayfield

Christof Schmitt

Barton Allen Smith

Stefan Ravizza

Martin G Kienzle

Ahmet Ekin

Ildar Khabibrakhmanov

Yashodhara M Patnaik

Frederik C M Kjeldsen

Sharon L Nunes

Jun Jun Li

Bernadette R Duponchel

Raymond A Cooke

Carlo Siebenschuh

Sharat Suresh Chikkerur

Holger Martens

Kristina Scherbaum

Xi Peng

Yuichi Fujiji

James Petterson

Thomas J Hammershoy

Craig Velliquette

Rasit Onur Topalogu

A framework is provided in which a querying agency can request (via a query entity) encrypted data through a service provider from a data owning agency that stores encrypted data. The framework uses homomorphic encryption. The data may be gallery entities, and each of the elements in the framework operate on doubly-encrypted information. The service provider compares a representation of an encrypted query entity from the querying agency and representations of encrypted gallery entities from the data owning agency, resulting in doubly-encrypted values of a metric between corresponding compared representations. The querying agency gets result(s), based on the metric, which indicate whether it is probable the service provider has data similar to or the same as query data in the query entity. The elements have to perform communication in order for the querying agency or the data owning agency to get cleartext information corresponding to the query entity.

Secure matching and identification of patterns

A method, apparatus and computer program product for homomorphic inference on a decision tree (DT) model. In lieu of HE-based inferencing on the decision tree, the inferencing instead is performed on a neural network (NN), which acts as a surrogate. To this end, the neural network is trained to learn DT decision boundaries, preferably without using the original DT model data training points. During training, a random data set is applied to the DT, and expected outputs are recorded. This random data set and the expected outputs are then used to train the neural network such that the outputs of the neural network match the outputs expected from applying the original data set to the DT. Preferably, the neural network has low depth, just a few layers. HE-based inferencing on the decision tree is done using HE inferencing on the shallow neural network. The latter is computationally-efficient and is carried without the need for bootstrapping.

Decision tree-based inference on homomorphically-encrypted data without bootstrapping

A neural network is configured to suppress an output of a mitigation node in a mitigation layer of the neural network. The neural network is pre-configured to recognize objects from inputs when operating using a processor and a memory. An actual input is sent to the neural network for object recognition, the actual input is an altered input. By suppressing the output of the mitigation node, the neural network is caused to avoid falsely recognizing an object from the actual input, where the altered input is configured to cause the neural network to falsely recognize the object from the actual input.

Mitigating false recognition of altered inputs in convolutional neural networks

Techniques regarding adaptive data recovery schemes 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 a data management component that can modify a data recovery scheme based on performance data exhibited by a network of data centers and a data recovery requirement. The data recovery scheme can direct a relocation of data within the network.

Data recovery modification based on performance data exhibited by a network of data centers and data recovery requirement

Generating visual data by defining a first action into a first set of objects and corresponding first set of motions, and defining a second action into a second set of objects and corresponding second set of motions. A relationship is then determined for the second action to the first action in terms of relationships between corresponding constituent objects and motions. Objects and motions are detected from visual data of first action. Visual data is composed for the second action from the data by transforming the constituent objects and motions detected in first action based on the corresponding determined relationships.

Augmentation for visual action data

A method makes container services persistently available. A computing device receives a request for implementation of a user job in a container environment, and assigns the user job to a compute runner agent of a plurality of compute runner agents to execute the user job. Each compute runner agent is associated with a compute container instance having a unique compute container identifier corresponding to the user job. A computing device assigns the user job to a balancer task to monitor progress of the user job, and assigns the user job to a storage agent to store artifacts associated with running the user job. A computing device receives a notification from the balancer task describing whether the runner agent is correctly running the user job. In response to the runner agent incorrectly running the user job, a computing device resurrects the user job in a new compute container instance.

Persistently available container services through resurrection of user jobs in new compute container instances designated as lead instances

Mechanisms are provided to provide an improved computer tool for determining and mitigating the presence of adversarial inputs to an image classification computing model. A machine learning computer model processes input data representing a first image to generate a first classification output. A cohort of second image(s), that are visually similar to the first image, is generated based on a comparison of visual characteristics of the first image to visual characteristics of images in an image repository. A cohort-based machine learning computer model processes the cohort of second image(s) to generate a second classification output and the first classification output is compared to the second classification output to determine if the first image is an adversarial image. In response to the first image being determined to be an adversarial image, a mitigation operation by a mitigation system is initiated.

Cohort based adversarial attack detection

In an approach for optimizing resources in a disaster recovery cleanup process, processors are configured for receiving transaction entries represented by transaction identifiers at a source database in communication with target databases via Synchronous-to-Asynchronous Traffic Converters (SATCs). Further, the processors are configured for transmitting a transaction payload from the SATCs to the target databases; identifying completed tracking entries corresponding to tracking entries having a complete status for the SATCs; deleting remaining transaction entries ranging from a transaction entry associated with a highest processed transaction identifier to a transaction entry associated with a lowest processed transaction identifier; providing a list of the remaining transaction entries that were deleted to predecessors of the SATCs; removing the remaining transaction entries from the SATCs if the transaction entries were delivered to all target databases; and detecting a topology change corresponding to one or more additional SATCs integrated with the one or more SATCs.

Optimizing resources in a disaster recovery cleanup process

This disclosure provides a method, apparatus and computer program product to create a full homomorphic encryption (FHE)-friendly machine learning model. The approach herein leverages a knowledge distillation framework wherein the FHE-friendly (student) ML model closely mimics the predictions of a more complex (teacher) model, wherein the teacher model is one that, relative to the student model, is more complex and that is pre-trained on large datasets. In the approach herein, the distillation framework uses the more complex teacher model to facilitate training of the FHE-friendly model, but using synthetically-generated training data in lieu of the original datasets used to train the teacher.

Depth-constrained knowledge distillation for inference on encrypted data

In some examples, a method of vector-raster data fusion includes receiving vector data for a geographical location, and statistically analyzing the vector data to obtain vector statistics. In some examples the method further includes rasterizing the vector statistics, and storing at least one of the vector data and the rasterized vector statistics together in a key-value store together with previously stored raster data for the geographical location. In some examples, the vector data further includes metadata, and the method further includes storing the metadata in at least one of the key-value store or a separate vector database.

Distributed vector-raster fusion

Methods and systems for generating a composite image in remote sensing applications are described. In an example, a device can receive an image having a plurality of points specified in an image space. The device can extract a portion of the image and transform points among the extracted portion from the image space to a parameter space defined by a distance parameter and an orientation parameter. The device can identify a set of intersection points in the parameter space that indicate at least one occurrence of a geometry feature in the extracted portion of the image. The device can augment the portion of the image with a plurality of new pixels based on the identified set of intersection points. The device can generate a composite image using the augmented image, where the composite image can include a plurality of augmented images corresponding to other portions of the image.

Scalable visual analytics for remote sensing applications

A technique for computationally-efficient privacy-preserving homomorphic inferencing against a decision tree. Inferencing is carried out by a server against encrypted data points provided by a client. Fully homomorphic computation is enabled with respect to the decision tree by intelligently configuring the tree and the real number-valued features that are applied to the tree. To that end, and to the extent the decision tree is unbalanced, the server first balances the tree. A cryptographic packing scheme is then applied to the balanced decision tree and, in particular, to one or more entries in at least one of: an encrypted feature set, and a threshold data set, that are to be used during the decision tree evaluation process. Upon receipt of an encrypted data point, homomorphic inferencing on the configured decision tree is performed using a highly-accurate approximation comparator, which implements a 'soft' membership recursive computation on real numbers, all in an oblivious manner.

Privacy-enhanced decision tree-based inference on homomorphically-encrypted data

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.

Predicting and correcting vegetation state

An AI model is trained by determining insights for a sequence of computations used in the AI model. The sequence is applied to encrypted data and label pair(s), wherein computational details of each of the computations are defined. Information may also be committed for selected ones of the sequence of computations into a distributed database. The committed information may include computational details used in processing performed for the selected computations, and the distributed database may have a property that the committed information for each selected computation is linked with a verifiable signature of integrity with a previously committed computation in the sequence. Indication is received from an end-user computer system of selected computation(s). Computational details of the indicated selected computation(s) are sent toward the end-user computer system for use by the end-user computer system for verifying the indicated selected computation(s). The end-user computer system can verify the selected computation(s).

Encrypted data model verification

Growing community of inventors

Fairfield County, CT, United States of America

Sharathchandra UmapathiRao Pankanti