International Business Machines Corporation

Globalfoundries Inc.

Ruchir Puri

Minsik Cho

Daniel Brand

Rajesh Ramkrishna Bordawekar

Hua Xiang

Haoxing Ren

David Shing-ki Kung

Ulrich Alfons Finkler

David Shing-Ki Kung

Louise Helen Trevillyan

Anthony Correale, Jr

Haifeng Qian

Matthew Mantell Ziegler

Zhigang Pan

Paul William Coteus

Luis Alfonso Lastras-Montano

Kerry Bernstein

Philip George Emma

Michele M Franceschini

Chin Ngai Sze

Brian R Konigsburg

Mark B Ritter

Emrah Acar

Livio B Soares

Mihir Choudhury

Rajiv V Joshi

Thomas Edward Rosser

Vincent Kulandaisamy

Andrew Augustus Bjorksten

Douglass T Lamb

David John Geiger

Smita Krishnaswamy

Daniel Joseph Friedman

Manish Anand Bhide

Stephen Victor Kosonocky

Mark Alan Lavin

Brian William Curran

Albert M Young

Allan Mark Hartstein

Sharad C Sundararajan

Jeannine Madelyn Trewhella

Ibrahim M Elfadel

Anthony DeGroff Drumm

Andrew James Sullivan

Victor N Kravets

David Wallach

Dorothy Kucar

Subhendu Roy

David James Hathaway

Charles Jay Alpert

Laertis Economikos

Hung Cai Ngo

Bruce Martin Fleischer

James Douglas Warnock

Stephen Edward Greco

Diptikalyan Saha

Wiren Dale Becker

Shyam Ramji

Joachim Keinert

Kush Raj Varshney

Prabhakar Nandavar Kudva

Tingdong Zhou

Adam P Matheny

Pin-Yu Chen

Amit Dhurandhar

Gary S Ditlow

Alvan Wing Ng

Pranay Kumar Lohia

Eli Arbel

Nathaniel Douglas Hieter

Gregory Allen Northrop

Karthikeyan Shanmugam

Soroush Abbaspour

Evanthia Papadopoulou

Ravi Chandra Chamarthy

Karthikeyan Natesan Ramamurthy

Xiaoping Tang

Chandramouli V Kashyap

Jagannathan Narasimhan

Subhrajit Bhattacharya

Michael D Amundson

Sean Michael Carey

Leon Stok

Mervyn Y Tan

Mohammed Fazil Fayaz

Rama Gopal Gandham

Pooja M Kotecha

David L Rude

Patricia A O'Neil

Veena S Pureswaran

Jeremy T Hopkins

Michael W Dotson

Tejaswini Pedapati

Reinaldo A Bergamaschi

Mark D Mayo

William R Migatz

John Mack Isakson

Atin Sood

Avinash Balakrishnan

Paul M Campbell

A batched Cholesky decomposition method, system, and non-transitory computer readable medium for a Graphics Processing Unit (GPU), include mirroring matrices to form paired matrices solving the paired matrices simultaneously.

Mirroring matrices for batched cholesky decomposition on a graphic processing unit

System, method, and recording medium for mirroring matrices for batched cholesky decomposition on a graphic processing unit

A post-processing method, system, and computer program product for post-hoc improvement of instance-level and group-level prediction metrics, including training a bias detector that learns to detect a sample that has an individual bias greater than a predetermined individual bias threshold value with constraints on a group bias, applying the bias detector on a run-time sample to select a biased sample in the run-time sample having a bias greater than the predetermined individual bias threshold bias value, and suggesting a de-biased prediction for the biased sample.

Post-hoc improvement of instance-level and group-level prediction metrics

A method, system, and computer program product, including generating a contrastive explanation for a decision of a classifier trained on structured data, highlighting an important feature that justifies the decision, and determining a minimal set of new values for features that alter the decision.

Model agnostic contrastive explanations for structured data

Mechanisms are provided for synthesizing a computer implemented neural network. An initially trained neural network is received and modified by introducing a new hidden layer of neurons and new connections that connect the new hidden layer of neurons to an output layer and a previous layer of neurons previously directly connected to the output layer of neurons to generate a modified neural network. The modified neural network is trained through one or more epochs of machine learning to generate modified weight values for the new connections and the new connections are pruned based on the modified weight values to remove a subset of the new connections and leaving remaining connections in the modified neural network. A merge operation is performed on the remaining connections in the modified neural network to generate a custom convolution filter and modified neural network. The modified neural network is then retrained for deployment.

Building of custom convolution filter for a neural network using an automated evolutionary process

Systems and methods for detecting indirect bias in machine learning models are provided. A computer-implemented method includes: receiving, by a computer device, a user request to detect transitive bias in a machine learning model; determining, by the computer device, correlations of attributes of neighboring data not included in a dataset of the machine learning model; ranking, by the computer device, the attributes based on the determined correlations; and returning, by the computer device, a list of the ranked attributes to a user that generated the user request.

Automatic correction of indirect bias in machine learning models

A batched Cholesky decomposition method, system, and non-transitory computer readable medium for a Graphics Processing Unit (GPU), include mirroring a second problem matrix of a second problem to a first problem matrix of a first problem as paired matrices and shifting the second problem matrix by N+1 and combining the first problem matrix and the mirrored second problem matrix into one matrix of (N+1)&#xd7;N, where the first problem shared memory comprises regular intervals, where the second problem shared memory is continuous, and where the GPU performs batched dense Cholesky decomposition with the one matrix from the combining to accelerate the Cholesky decomposition.

System, method, and recording medium for mirroring matrices for batched Cholesky decomposition on a graphic processing unit

Mechanisms, in a system comprising a host system and at least one accelerator device, for performing a concept analysis operation are provided. The host system extracts a set of one or more concepts from an information source and provides the set of one or more concepts to the accelerator device. The host system also provides at least one matrix representation data structure representing a graph of concepts and relationships between concepts in a corpus. The accelerator device executes the concept analysis operation internal to the accelerator device to generate an output vector identifying concepts in the corpus, identified in the at least one matrix representation data structure, related to the set of one or more concepts extracted from the information source. The accelerator device outputs the output vector to the host system which utilizes the output vector to respond to a request submitted to the host system associated with the information source.

Concept analysis operations utilizing accelerators

Apparatuses and Methods for sorting a data set. A data storage is divided into a plurality of buckets that is each associated with a respective key value. A plurality of stripes is identified in each bucket. A plurality of data stripe sets is defined that has one stripe within each respective bucket. A first and a second in-place partial bucket radix sort are performed on data items contained within the first and second data stripe sets, respectively, using an initial radix. Incorrectly sorted data items in the first bucket are grouped by a first processor and incorrectly sorted data items in the second bucket are grouped by a second processor into a respective incorrect data item group within each bucket. A radix sort is then performed using the initial radix on the items within the respective incorrect data item group. A first level sorted output is produced.

Parallelized in-place radix sorting

An iterative graph algorithm accelerating method, system, and computer program product, include recording an order of access nodes in a memory layout, reordering the access nodes in the memory layout in accordance with the recorded order, and updating edge information of the reordered access nodes.

System, method and computer program product for accelerating iterative graph algorithms by memory layout optimization

An information processing system, computer readable storage medium, and method for accelerated radix sort processing of data elements in an array in memory. The information processing system stores an array of data elements in a buffer memory in an application specific integrated circuit radix sort accelerator. The array has a head end and a tail end. The system radix sort processing, with a head processor, data elements starting at the head end of the array and progressively advancing radix sort processing data elements toward the tail end of the array. The system radix sort processing, with a tail processor, data elements starting at the tail end of the array and progressively advancing radix sort processing data elements toward the head end of the array, the tail processor radix sort processing data elements in the array contemporaneously with the head processor radix sort processing data elements in the array.

Radix sort acceleration using custom asic

A first quicksort is performed in parallel across pairs of partitions of a dataset assigned to respective ones of available processors, including swapping elements of a first partition of a given one of the pairs that are larger than a pivot with elements of a second partition of the given pair that are smaller than the pivot. A second quicksort is performed in parallel across those partitions having elements left unsorted by the first quicksort, and first misplaced elements from a first side of the dataset corresponding to the first partition are swapped with second misplaced elements from a second side of the dataset corresponding to the second partition to produce a first dataset having elements equal to or lower than the pivot and a second dataset having elements equal to or higher than the pivot.

Parallel quicksort

A batched Cholesky decomposition method, system, and non-transitory computer readable medium for a Graphics Processing Unit (GPU), include mirroring a second problem matrix of a second problem to a first problem matrix of a first problem as paired matrices and shifting the second problem matrix by N+1 and combining the first problem matrix and the mirrored second problem matrix into one matrix of (N+1)&#xd7;N by merging the first problem matrix and the mirrored second problem matrix. The first problem matrix and the second problem matrix are symmetric and positive definite matrices.

System, Method, and recording medium for mirroring matrices for batched Cholesky decomposition on a graphic processing unit

A batched Cholesky decomposition method, system, and non-transitory computer readable medium for a Graphics Processing Unit (GPU), include mirroring a second problem matrix of a second problem to a first problem matrix of a first problem as paired matrices and shifting the second problem by N+1, combining the first problem matrix and the mirrored second problem matrix into one matrix of (N+1)&#xd7;N, and reading the fixed size data length of the one square matrix with a fixed data interval for both the first problem and the second problem.

Growing community of inventors

Baldwin Place, NY, United States of America

Ruchir Puri