362 research outputs found
Responsible ML Datasets
No full textIn this study, we discuss the importance of Responsible Machine Learning Datasets through the lens of fairness, privacy, and regulatory compliance and present a large audit of Computer Vision datasets. The audit is conducted through evaluation of the proposed responsible rubric. After surveying over 100 datasets, our detailed analysis of 60 distinct datasets highlights a universal susceptibility to fairness, privacy, and regulatory compliance issues.
Please cite the paper below.
Mittal, Surbhi, Kartik Thakral, Richa Singh, Mayank Vatsa, Tamar Glaser, Cristian Canton Ferrer, Tal Hassner. "On Responsible Machine Learning Datasets Emphasizing Fairness Privacy and Regulatory Norms with Examples in Biometrics and Healthcare." Nature Machine Intelligence (2024).
@article{mittal2024responsible,
title={On Responsible Machine Learning Datasets Emphasizing Fairness Privacy and Regulatory Norms with Examples in Biometrics and Healthcare},
author={Mittal, Surbhi, and Thakral, Kartik and Singh, Richa and Vatsa, Mayank and Glaser, Tamar and Ferrer, Cristian Canton and Hassner, Tal},
journal={Nature Machine Intelligence},
year={2024},
publisher={Nature Publishing Group UK London}
Responsible ML Datasets
No full textIn this study, we discuss the importance of Responsible Machine Learning Datasets through the lens of fairness, privacy, and regulatory compliance and present a large audit of Computer Vision datasets. The audit is conducted through evaluation of the proposed responsible rubric. After surveying over 100 datasets, our detailed analysis of 60 distinct datasets highlights a universal susceptibility to fairness, privacy, and regulatory compliance issues.
Please cite the paper below.
Mittal, Surbhi, Kartik Thakral, Richa Singh, Mayank Vatsa, Tamar Glaser, Cristian Canton Ferrer, Tal Hassner. "On Responsible Machine Learning Datasets Emphasizing Fairness Privacy and Regulatory Norms with Examples in Biometrics and Healthcare." Nature Machine Intelligence (2024).
@article{mittal2024responsible,
title={On Responsible Machine Learning Datasets Emphasizing Fairness Privacy and Regulatory Norms with Examples in Biometrics and Healthcare},
author={Mittal, Surbhi, and Thakral, Kartik and Singh, Richa and Vatsa, Mayank and Glaser, Tamar and Ferrer, Cristian Canton and Hassner, Tal},
journal={Nature Machine Intelligence},
year={2024},
publisher={Nature Publishing Group UK London}
Approximation of Signals (Functions) by Trigonometric Polynomials in Lp-Norm
No full textMittal and Rhoades (1999, 2000) and Mittal et al. (2011) have initiated a study of error estimates En(f) through
trigonometric-Fourier approximation (tfa) for the situations in which the summability matrix T does not have monotone rows. In this paper, the first author continues the work in the direction for T to be a Np-matrix. We extend two theorems on summability matrix Np of Deger et al. (2012) where they have extended two theorems of Chandra (2002) using Cλ-method obtained by deleting a set of rows from Cesàro matrix C1. Our theorems also generalize two theorems of Leindler (2005) to Np-matrix which in turn generalize the result of Chandra (2002) and Quade (1937)
Highly scalable solution of incompressible Navier-Stokes equations using the spectral element method with overlapping grids
No full textWe present a highly-flexible Schwarz overlapping framework for simulating turbulent fluid/thermal transport in complex domains. The approach is based on a variant of the Schwarz alternating method in which the solution is advanced in parallel in separate overlapping subdomains. In each domain, the governing equations are discretized with an efficient high-order spectral element method (SEM). At each step, subdomain boundary data are determined by interpolating from the overlapping region of adjacent subdomains. The data are either lagged in time or extrapolated to higher-order temporal accuracy using a novel stabilized predictor-corrector algorithm. Matrix stability analysis is used to determine the optimal number of corrector iterations. Stability and accuracy are further improved with an optimal mass flux correction to guarantee mass conservation throughout the domain. The method supports an arbitrary number of subdomains. A new multirate time-stepping scheme is developed (a first for incompressible flow simulations) that allows the underlying equations to be advanced with time-step sizes varying as much as an order-of-magnitude between adjacent domains. All the developments maintain the third-order temporal convergence and exponential convergence of the originating SEM framework. This dissertation also presents a mesh optimizer that has been specifically designed for meshes generated for turbulent flow problems. The optimizer supports surface mesh improvement, which minimizes geometrical approximation errors. The smoother is shown to reduce the computational cost of numerical calculations by as much as 40%. Numerous examples illustrate the effectiveness of these new technologies for analyzing challenging turbulence problems that were previously infeasible.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-12-01The student, Ketan Mittal, accepted the attached license on 2019-10-07 at 11:58.The student, Ketan Mittal, submitted this Dissertation for approval on 2019-10-07 at 12:08.This Dissertation was approved for publication on 2019-10-09 at 15:36.DSpace SAF Submission Ingestion Package generated from Vireo submission #14486 on 2020-02-28 at 17:20:54Made available in DSpace on 2020-03-02T22:12:10Z (GMT). No. of bitstreams: 2
MITTAL-DISSERTATION-2019.pdf: 43195402 bytes, checksum: ee2355b57595dfdb6f0483c839c4b9ce (MD5)
LICENSE.txt: 4209 bytes, checksum: db35dd4507774e72a4ad849ff9b8751d (MD5)
Previous issue date: 2019-10-09Embargo set by: Seth Robbins for item 113863
Lift date: 2022-03-02T22:12:26Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 113863
Lift date: 2022-03-02T22:15:21Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 113863
Lift date: 2022-03-02T22:18:25Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemOpen Restriction set for Item 113863 on 2020-03-04T16:22:37Z with date null by [email protected] Restriction set for Item 113863 on 2020-03-04T16:22:39Z with date null by [email protected]
Application of SDP to product rules and quantum query complexity
No full textIn recent years, semidefinite programming has played a vital role in shaping complexity theory and quantum computing. There have been numerous applications ranging from estimating quantum values, over approximating combinatorial quantities, to proving various bounds. This work extends the use of semidefinite programs (SDPs) to proving product rules and to characterizing quantum query complexity. In the first application, we provide a general framework to establishing product rules for quantities that can be expressed (or approximated) using SDPs. We use duality theory to give product rules, which bound the value of the ``product'' of two problems in terms of their value. Some previous results have implicitly used the properties of SDPs to give such product rules. Here we give sufficient and necessary conditions under which these approaches work, thereby enabling us to capture these previous results under our unified framework. We also include a discussion about alternate definitions of what a ``product'' means and how they fit into our approach. The second application provides an SDP characterization of quantum query complexity, which is one of the ways in which complexity of a function can be measured. It is known that quantum query complexity can be lower bounded by the so-called ``adversary method'' which is expressible as a semidefinite program. Recently, Ben Reichardt showed that the adversary method leads to a tight lower bound for boolean functions by converting the solution of this SDP (of adversary method) into an algorithm. We show that a related SDP, called ``witness size'' in this thesis, provides a tight bound on the quantum query complexity of non boolean functions (total as well as partial). This witness size SDP is also used to give composition results for quantum query complexity. We also show that the witness size is bounded by a constant multiple of the adversary bound. Finally, we briefly explore whether other convex programming paradigms can be useful in complexity theory. One of them is copositive programming. We show that one of the recent result about parallel repetition of unique games, by Barak et.al., can be interpreted as an application of copositive programming.Ph.D.Includes bibliographical referencesIncludes vitaby Rajat Mitta
Efficient sequential decision-making algorithms for container inspection operations
No full textSequential diagnosis is an old subject, but one that has become increasingly important recently. There exists a need for new models and algorithms as the traditional methods for making decisions sequentially do not scale. Motivated by the problem of container inspection at the U.S. ports, we investigate the problem of finding efficient algorithms for sequential diagnosis. More specifically, we formulate the port of entry inspection sequencing task as a problem of finding an optimal binary decision tree for an appropriate Boolean decision function. We provide new algorithms that are computationally more efficient than those previously presented by Stroud and Saeger [31] and Anand et al [1]. We achieve these efficiencies through a combination of specific numerical methods for finding optimal thresholds for sensor functions and two novel binary decision tree search algorithms that operate on a space of potentially acceptable binary decision trees. The improvements enable us to analyze substantially larger applications than was previously possible.
We try to solve the problem of finding an optimal inspection strategy by breaking it into two sub-problems - 1. Finding sensor threshold values that minimize the cost for a given binary decision tree and 2. ``Searching'' for the cheapest binary decision tree in a large space of trees or equivalence classes of trees. For solving the first problem, we explore various standard non-linear optimization techniques and also propose a novel algorithm by combining the gradient descent method and Newton's method in optimization to compute optimal thresholds for any given tree. We propose two novel search algorithms - A stochastic search method and a genetic algorithms based search method, as a solution to the second sub-problem. We also propose ``neighborhood'' operations to move from one tree to another in the proposed tree space and prove that the tree space is irreducible under these neighborhood operations.
We report results from numerous experiments with and without imposing restrictions on the tree space and examine how the optimal binary decision trees vary with these changes. For example, for most of the work in this thesis, we restrict the tree space to constitute only ``complete'' and ``monotonic'' binary decision trees. Later, we ``shrink'' the tree space by discovering equivalence classes of trees while we ``expand'' the tree space by removing the monotonicity constraint.M.S.Includes bibliographical references (p. 61-63)
Regional empty marine container management
No full textEmpty container repositioning is one of the longstanding and ongoing issues in the containerized maritime trade. Even though it is a non-revenue generating, expensive and undesirable exercise, it is an integral part of an overall efficient global transportation system, which balances demand and supply of empty containers between regions. Empty containers are repositioned at three levels - global, inter-regional and regional-level. The focus of this dissertation is at the regional level of empty container repositioning.
Regional repositioning of empty containers involves empty container movement between regional importers, marine terminals, empty container depots, and export customers. This chain movement generates excessive unproductive empty vehicle miles in a region. The problem of empty vehicle miles travelled becomes more prominent when empty container depots are located close to the port and import and export customers are inland. Stakeholders incur large system costs in repositioning empty containers between the regional import-export business locations and the port/depots. Regions with high import activity are concerned with the increase in containerized trade volumes and the persistent trade imbalance because of the capacity shortfall at their existing depots.
This thesis addresses the above two regional concerns of excessive empty vehicle miles and empty container storage capacity shortfall by proposing an 'Inland-Depots-for-Empty-Containers (IDEC)' system. It recommends opening new empty container depots inland in the region, closer to high volume import-export customer clusters, in addition to the depots currently being located near the ports. The dissertation discusses the feasibility, viability, and effectiveness of the proposed system.
It develops mathematical models for the IDEC system to determine the optimal number and location of inland depots in a given region under deterministic and stochastic demand patterns. Exploiting the structure of the NP-hard problem, it develops a heuristic based on the randomized rounding algorithm to solve large scale, realistic depot-location problems. To implement a successful and sustainable IDEC system, it explicitly considers the varied perspectives of different maritime stakeholders involved in the container movement. Based on the models and quantitative analyses, it demonstrates that an IDEC system has great potential in improving regional empty moves, increasing both business profitability and social welfare simultaneously.Ph.D.Includes bibliographical references (p. 133-144)
Breast cancer diagnosis using Fourier transform infrared imaging and statistical learning
No full textCancer alters both the morphological and the biochemical properties of multiple cell types in a tissue. Generally, the morphology of epithelial cells is practically used for routine disease diagnoses. Current histopathological diagnosis involves manual interpretation of stained images for patient diagnosis. This is prone to inter- observer variability leading to low concordance rates amongst pathologists. Further, since structural features are mostly just defined for epithelial alterations during tumor progression, the use of associated stromal changes is limited. To overcome these challenges, digital analysis of these images is suggested that can result in the determination of precise and quantitative metrics both for epithelial and stromal disease signatures.
In my dissertation work, I focused on building combinatorial approaches using chemical imaging, histopathology images, machine learning and deep learning. An emerging area of investigation is using spectrometry to perform tissue analysis that utilizes chemical imaging coupled to machine learning to identify spectral signatures indicative of disease state and its progression. Infrared spectroscopic imaging biochemically characterizes breast cancer, both for the epithelial cells and the tumor-associated microenvironment. I utilized multiple breast tissue assignments and a supervised learning approach to create different histologic and pathologic models using both high definition (HD) and standard definition (SD) data. The comparison of HD and SD modalities shows that new information richness associated with better spatial resolution facilitates the creation of complex, multiclass models of breast tissue without compromising on the sensitivity and the specificity of tissue segmentation. These models were then extended to discrete frequency measurements for rapid analysis cutting down tissue analysis time from days to minutes, making the technology feasible for research optimizations and clinical translation. Additionally, I optimized and tuned existing convolutional neural networks to identify different disease states in breast cancer and the corresponding microenvironment. Finally, I developed analytical tools for early detection and standardized analysis of stained image data. This can offer new opportunities for objective, accurate and comprehensive patient diagnosis and prognostics.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2021-08-01The student, Shachi Mittal, accepted the attached license on 2019-07-10 at 09:35.The student, Shachi Mittal, submitted this Dissertation for approval on 2019-07-10 at 15:26.This Dissertation was approved for publication on 2019-07-11 at 13:29.DSpace SAF Submission Ingestion Package generated from Vireo submission #14241 on 2019-11-26 at 14:03:23Made available in DSpace on 2019-11-26T20:59:42Z (GMT). No. of bitstreams: 2
MITTAL-DISSERTATION-2019.pdf: 6909092 bytes, checksum: 289f8ec7d3af773587c848c03be374ee (MD5)
LICENSE.txt: 4210 bytes, checksum: 282484ffe017e9cf19ac33938290f8bc (MD5)
Previous issue date: 2019-07-11Embargo set by: Seth Robbins for item 113072
Lift date: 2021-11-26T20:59:54Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction set for Item 113072 on 2021-07-09T15:16:01Z with date 2023-11-26 by [email protected] Restriction set for Item 113072 on 2021-07-09T15:16:05Z with date 2023-11-26 by [email protected] submitted a closed access extension request that was approved by the Thesis Office.Limite
Spectral element mesh generation and improvement methods
No full textMade available in DSpace on 2017-03-01T16:37:06Z (GMT). No. of bitstreams: 2
MITTAL-THESIS-2016.pdf: 6322867 bytes, checksum: ceeadf110747990a0a9d8d4edfd97f89 (MD5)
LICENSE.txt: 4209 bytes, checksum: e320f9ba4f501ad2c60568d7212152e0 (MD5)
Previous issue date: 2016-12-06Embargo set by: Seth Robbins for item 98625
Lift date: 2019-03-01T16:37:19Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 98625 on 2019-03-02T10:15:21Z.Meshing tools for finite element meshes have been studied extensively over the last few decades. However, relatively less attention has been paid to spectral element meshes. This thesis focuses on mesh generation and mesh improvement methods for spectral element meshes. A mesh smoother, based on a combination of Laplacian smoothing and optimization, has been developed and implemented in Nek5000, an open-source spectral element method based incompressible flow solver. The smoother takes a valid mesh as an input and outputs an improved mesh. Comparison of the original and smoothed mesh has shown that mesh smoothing decreases the iteration count of iterative solvers. This reduction is anticipated from an observed decrease in the ratio of the maximum to minimum eigenvalues of the upper Hessenberg matrix generated by the GMRES method. The mesh smoother was tested on various meshes for complicated geometries, and was found to improve the computational efficiency of calculations by up to 20% which is helping save 100,000s of cpu-hours on high-performance computing machines. A mesh skinning tool has also been developed which adds boundary layer resolving elements of user-specified thickness at user-specified surfaces in an existing mesh. This translates into savings in terms of human time and effort since the user can now robustly add boundary layer resolving elements instead of manually meshing the geometry to add these elements. Additionally, tools have been developed that generate meshes for geometries like turbine blades and random-array of cylinders (to simulate flow in vegetated channels), in a matter of seconds. Finally, a tetrahedral (tet) to hexahedron (hex) mesh converter has been implemented, that generates spectral element meshes for any complicated geometry by taking an all-tet mesh and converting it to an all-hex spectral element mesh. This tool has been developed to quickly generate all-hex meshes with minimal user intervention.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2018-12-01The student, Ketan Mittal, accepted the attached license on 2016-12-05 at 11:26.The student, Ketan Mittal, submitted this Thesis for approval on 2016-12-06 at 10:16.This Thesis was approved for publication on 2016-12-06 at 16:34.DSpace SAF Submission Ingestion Package generated from Vireo submission #10431 on 2017-02-28 at 14:37:1
Dark energy model in higher-dimensional FRW universe with respect to generalized entropy of Sharma and Mittal of flat FRW space–time
No full textIn this study, we report the state parameter of dark energy in higher dimensional Friedmann–Robertson–Walker (FRW) space–time according to generalized entropy of Sharma and Mittal. In this case we analyze the state parameter of dark energy according to today’s observational evidence.The accepted manuscript in pdf format is listed with the files at the bottom of this page. The presentation of the authors' names and (or) special characters in the title of the manuscript may differ slightly between what is listed on this page and what is listed in the pdf file of the accepted manuscript; that in the pdf file of the accepted manuscript is what was submitted by the author
- …
