International Business Machines Corporation

Arizona State University

Harvard College

Karthikeyan Natesan Ramamurthy

Sharathchandra UmapathiRao Pankanti

Aleksandr Y Aravkin

Chung-Ching Lin

Dennis Wei

John Richard Smith

Kush Raj Varshney

Andreas Spanias

Jayaraman Jayaraman Thiagarajan

David H Frakes

Manish Anand Bhide

Ruchir Puri

Diptikalyan Saha

Elizabeth Daly

Supriyo Chakraborty

Jun Li Wang

Rahul Nair

Oznur Alkan

Aleksandra Mojsilovic

Pin-Yu Chen

Pranay Kumar Lohia

Payel Das

Shilpa N Mahatma

Flavio Du Pin Calmon

Massimiliano Mattetti

Skyler Speakman

Kevin N Tran

Gigi Yuen-Reed

Jingwei Yang

Rachita Chandra

Dongping Fang

John H Bauer

Raphael Viguier

Prasanna Sattigeri

Amanda Coston

Zairah Mustahsan

Pu Zhao

Embodiments are disclosed for a method. The method includes receiving feedback decision rules for multiple predictions by a trained machine learning model. generating a feedback rule set based on the feedback decision rules. The method further includes generating an updated training dataset based on an original training dataset and an updated feedback rule set. The updated feedback rule set resolves one or more conflicts of the feedback rule set, and the updated training dataset is configured to train the machine learning model to move a decision boundary. Generating the updated training dataset includes generating multiple updated training instances by applying one of the feedback decision rules to a training instance of the original training dataset.

Moving decision boundaries in machine learning models

Obtain, from an existing machine learning classifier, original probabilistic scores classifying samples taken from two or more groups into two or more classes via supervised machine learning. Associate the original probabilistic scores with a plurality of original Lagrange multipliers. Adjust values of the plurality of original Lagrange multipliers via low-dimensional convex optimization to obtain updated Lagrange multipliers that satisfy fairness constraints as compared to the original Lagrange multipliers. Based on the updated Lagrange multipliers, closed-form transform the original probabilistic scores into transformed probabilistic scores that satisfy the fairness constraints while minimizing loss in utility. The fairness constraints are with respect to the two or more groups.

Optimized score transformation for fair classification

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

Techniques for sanitization of machine learning (ML) models are provided. A first ML model is received, along with clean training data. A path is trained between the first ML model and a second ML model using the clean training data. A sanitized ML model is generated based on at least one point on the trained path. One or more ML functionalities are then facilitated using the sanitized ML model.

Mitigating adversarial effects in machine learning systems

A method of utilizing a computing device to correct source data used in machine learning includes receiving, by the computing device, first data. The computing device corrects the source data via an application of a covariate shift to the source data based upon the first data where the covariate shift re-weighs the source data.

Enhancing fairness in transfer learning for machine learning models with missing protected attributes in source or target domains

A method and system of stitching a plurality of image views of a scene, including grouping matched points of interest in a plurality of groups, and determining a similarity transformation with smallest rotation angle for each grouping of the matched points. The method further includes generating virtual matching points on non-overlapping area of the plurality of image views and generating virtual matching points on overlapping area for each of the plurality of image views.

System and method for perspective preserving stitching and summarizing views

Modularized data processing systems and methods for its use are provided. Processing a current job can reuse data generated for a previously processed job to the extent the two share parameter configurations. Similarly, outputs of processing modules generated during processing the previously processed job can be used as inputs to processing modules processing a current job, if the two jobs share some parameter configurations.

Job management in data processing system

A method includes performing contextual association of entities using multi-source data. For each context the method performs co-clustering to identify distinct expert-skill associations; constructing single-entity unipartite graph representations and performing a random walk within each single-entity unipartite graph; for each single-entity unipartite graph, obtaining steady state distributions using the random walks to obtain clusters of experts and skills; performing a weighted two-way random walk across entity graphs (graph edges), giving preference to traversal within members of the same co-cluster; and performing link prediction for each context by dynamically adding edges, and obtaining overall skills predictions, analyses and inferences by merging the contexts and weighting the links of each context. The method can also use the context-specific weights obtained from the co-association information in a matrix completion procedure, and finally merge the context-specific outputs to obtain overall skills predictions, analyses and inferences. A computer program product and a system are also disclosed for performing the method.

Method, system and computer program product for automating expertise management using social and enterprise data

A method includes obtaining two-dimensional measurements of a given scene from sensors, the given scene comprising a plurality of scene points, distributing the given scene into bundles each associated with at least one sensor and at least one scene point, establishing constraints associated with the sensors and the plurality of scene points that are associated with two different bundles, estimating, individually for each of the bundles, a set of parameters for the at least one sensor and the at least one scene point associated with that bundle utilizing at least one of the two-dimensional measurements, generating a consensus parameter set for the one or more sensors and the plurality of scene points based on the estimated parameter sets, the established constraints being utilized to reconcile differences in parameter values in the estimated parameter sets, and producing a three-dimensional representation of the given scene utilizing the consensus parameter set.

Distributed processing for producing three-dimensional reconstructions

Methods and systems for recovering corrupted/degraded images using approximations obtained from an ensemble of multiple sparse models are disclosed. Sparse models may represent images parsimoniously using elementary patterns from a 'dictionary' matrix. Various embodiments of the present disclosure involve simple and computationally efficient dictionary design approach along with low-complexity reconstruction procedure that may use a parallel-friendly table-lookup process. Multiple dictionaries in an ensemble model may be inferred sequentially using greedy forward-selection approach and can incorporate bagging/boosting strategies, taking into account application-specific degradation. Recovery performance obtained using the proposed approaches with image super resolution and compressive recovery can be comparable to or better than existing sparse modeling based approaches, at reduced computational complexity. By including ensemble models in hierarchical multilevel learning, where multiple dictionaries are inferred in each level, further performance improvements can be obtained in image recovery, without significant increase in computational complexity.

Ensemble sparse models for image analysis and restoration

Measuring the number of glomeruli in the entire, intact kidney using non-destructive techniques is of immense importance in studying several renal and systemic diseases. In particular, a recent Magnetic Resonance Imaging (MRI) technique, based on injection of a contrast agent, cationic ferritin, has been effective in identifying glomerular regions in the kidney. In various embodiments, a low-complexity, high accuracy method for obtaining the glomerular count from such kidney MRI images is described. This method employs a patch-based approach for identifying a low-dimensional embedding that enables the separation of glomeruli regions from the rest. By using only a few images marked by the expert for learning the model, the method provides an accurate estimate of the glomerular number for any kidney image obtained with the contrast agent. In addition, the implementation of our method shows that this method is near real-time, and can process about 5 images per second.

Measuring glomerular number from kidney MRI images

A robust method to automatically segment and identify tumor regions in medical images is extremely valuable for clinical diagnosis and disease modeling. In various embodiments, an efficient algorithm uses sparse models in feature spaces to identify pixels belonging to tumorous regions. By fusing both intensity and spatial location information of the pixels, this technique can automatically localize tumor regions without user intervention. Using a few expert-segmented training images, a sparse coding-based classifier is learned. For a new test image, the sparse code obtained from every pixel is tested with the classifier to determine if it belongs to a tumor region. Particular embodiments also provide a highly accurate, low-complexity procedure for cases when the user can provide an initial estimate of the tumor in a test image.

Kernel sparse models for automated tumor segmentation

A method and a system of stitching a plurality of views of a scene, the method including extracting points of interest in each view to include a point set from each of the plurality of image views of the scene, matching the points of interest and reducing an outlier, grouping the matched points of interest in a plurality of groups, determining a similarity transformation with a smallest rotation angle for each grouping of the match points, generating virtual matching points on a non-overlapping area of the plurality of image views, generating virtual matching points on an overlapping area for each of the plurality of image views, and calculating piecewise projective transformations for the plurality of image views.

Growing community of inventors

Yorktown Heights, NY, United States of America

Karthikeyan Natesan Ramamurthy