154 research outputs found
Interaction of a railway tunnel with a deep slow landslide in clay shales
No full textThe Varco d’Izzo landslide system (Basilicata Region, Italy) develops at the suburbs of the city of Potenza, capital of the region, and is crossed by two transport infrastructures of local importance: the national highway Basentana and the national railway line. This paper is focused on the effects of slope movements on the railway tunnel which was built in the accumulation of an earthflow of the landslide system. The earthflow displacements were slow but continuous in the monitoring period 2005-2015 and in the order of one to several cm/year. They have led, not far from the railway tunnel area, to the eviction of a house, the dismantling of a pedestrian bridge, damages to roads and other structures. The tunnel was completely re-built in 1992 between two rows of piles, by the cut-and-cover method, after the previous tunnel had suffered severe damage due to the landslide. Available inclinometer data seem to suggest that, locally, the tunnel with its piles is hindering landslide displacements. In fact, measurements carried out in the vicinity of the tunnel, upslope from it, do not show a slip surface crossing the piles. On the other hand, landslide displacements are observed both farther, upslope from the tunnel, and downslope from it. The resultant of earth pressures acting on the tunnel is thus, probably, increasing with time. The distribution of landslide displacements around the tunnel until recent years is herein analyzed. Results of site surveys are reported. The causes of the current state of deformation of the tunnel, which was evaluated by laserscanning, are examined with the help of simplified calculations and FEM simulations
Topology-aware distributed graph processing for tightly-coupled clusters
Full text linkCloud applications have burgeoned over the last few years, but they are typically written for loosely-coupled clusters such as datacenters. In this thesis we investigate how one can run cloud applications in tightly-coupled clusters and network topologies, namely super-computers. Specifically, we look at a class of distributed machine learning systems called distributed graph processing systems, and run them on NCSA Blue Waters. Partitioning the graph is key to achieving performance in distributed graph processing systems. We present new topology-aware partitioning techniques that better exploit the structure of the network topologies in supercomputers. Compared to existing work, our new Restricted Oblivious and Grid Centroid partitioning approaches produce 25-33% improvement in makespan, along with a sizable reduction in network traffic. We also discuss optimizations such as smart network buffers that further amplify the improvement. To help operators select the best graph partitioning technique, we culminate our experimental results into a decision tree.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2020-05-01The student, Mayank Bhatt, accepted the attached license on 2018-04-23 at 17:13.The student, Mayank Bhatt, submitted this Thesis for approval on 2018-04-23 at 17:20.This Thesis was approved for publication on 2018-04-24 at 15:21.DSpace SAF Submission Ingestion Package generated from Vireo submission #12435 on 2018-08-31 at 17:21:19Made available in DSpace on 2018-09-04T20:36:52Z (GMT). No. of bitstreams: 2
BHATT-THESIS-2018.pdf: 1415794 bytes, checksum: e08311d8168967b2e47baf1ef67f7fdc (MD5)
LICENSE.txt: 4209 bytes, checksum: b810a770b0873fc45062dd7e9ce83fde (MD5)
Previous issue date: 2018-04-24Embargo set by: Seth Robbins for item 107297
Lift date: 2020-09-04T20:37:00Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 107297
Lift date: 2020-09-04T20:42:08Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 107297 on 2020-09-05T09:15:32Z
The Self-Destructive Nature of Human Divisions: When One Species Begins to See Itself as Many
No full textThis paper explores the paradoxical nature of human civilization, which, despite belonging to a single biological species (Homo sapiens), remains fundamentally fragmented by psychological, political, and ideological divisions. Author Mayank Singh argues that these constructed identities—ranging from religion and nationality to economic class and political ideology—are not harmless cultural differences but powerful structures that shape perception and foster systemic hostility. These divisions are largely maintained through unconscious conditioning and inherited social structures rather than rational choice.
The text highlights a critical existential risk: as modern humanity wields technologies capable of global destruction (such as nuclear weapons and AI), the continued reliance on "us versus them" mentalities threatens the survival of both the human species and the global biosphere. Singh posits that these rigid separations move civilization away from the interconnected patterns found in nature. Ultimately, the paper suggests that the future of civilization depends on a profound transformation in consciousness—transcending constructed identities to recognize a shared biological and cosmic existence before self-destructive logic leads to terminal conflict
Seeing Life as It Is: Beyond Human-Centered Existence
No full textThis paper critiques the anthropocentric foundations of modern civilization, arguing that human progress is frequently achieved through the systematic exploitation and destruction of non-human life. The author, Mayank Singh, explores the "illusion of human centrality," wherein animals and ecosystems are reduced to commodities, luxury products, or secondary participants in the planet's life system. A significant focus is placed on "conditioned demand"—consumption driven by luxury and convenience rather than survival—and the "economy of cruelty" that sustains it.
The work further identifies a contradiction in "symbolic respect," noting that many societies grant sacred status to specific animals based on religious mythology while remaining indifferent to the industrial-scale suffering of others. Singh proposes a shift toward "clear perception," a secular and non-utilitarian framework that recognizes all organisms as equivalent living processes. Ultimately, the paper suggests that a truly conscious civilization must redefine progress not through economic or technological growth, but through its capacity for ecological coexistence and respect for the full diversity of life
PoliWAM: An Exploration of a Large Scale Corpus of Political Discussions on WhatsApp Messenger
No full textWhatsApp Messenger is one of the most popular channels for spreading information with a current reach of more than 180 countries and 2 billion people. Its widespread usage has made it one of the most popular media for information propagation among masses during any socially engaging event. In the recent past, several countries have witnessed its effectiveness and influence in political and social campaigns. We observe a high surge in information and propaganda flow during elections. To explore such activities, in this paper, we discuss challenges, methodology, and opportunities in data curation from WhatsApp for politics-based exploratory studies. As a use case, we study the period before, during, and after the Indian General Elections 2019, encompassing all major Indian political parties. We present several complementing insights into the investigative and sensational news stories from the same period. Exploratory data analysis and experiments showcase several exciting results and future research opportunities. To facilitate reproducible research, we make the anonymized datasets available in the public domain.
If you are using this dataset as part of your research, please cite the following paper
@article{srivastava2020poliwam,
title={PoliWAM: an exploration of a large scale corpus of political discussions on WhatsApp messenger},
author={Srivastava, Vivek and Singh, Mayank},
journal={arXiv preprint arXiv:2010.13263},
year={2020}
Studies in program obfuscation
No full textThesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mathematics, 2010.This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.Cataloged from student submitted PDF version of thesis.Includes bibliographical references (p. 159-164).Program obfuscation is the software analog to the problem of tamper-proofing hardware. The goal of program obfuscation is to construct a compiler, called an "obfuscator," that garbles the code of a computer program while maintaining its functionality. Commercial products exist to perform this procedure, but they do not provide a rigorous security guarantee. Over the past decade, program obfuscation has been studied by the theoretical cryptography community, where rigorous definitions of security have been proposed and obfuscators have been constructed for some families of programs. This thesis presents three contributions based on the virtual black-box security definition of Barak et al [10]. First, we show tight connections between obfuscation and symmetric-key encryption. Specifically, obfuscation can be used to construct an encryption scheme with strong leakage resilience and key-dependent message security. The converse is also true, and these connections scale with the level of security desired. As a result, the known constructions and impossibility results for each primitive carry over to the other. Second, we present two new security definitions that augment the virtual black-box property to incorporate non-malleability. The virtual black-box definition does not prevent an adversary from modifying an obfuscated program intelligently. By contrast, our new definitions provide software with the same security guarantees as tamper-proof and tamper-evident hardware, respectively. The first definition prohibits tampering, and the second definition requires that tampering is detectable after the fact. We construct non-malleable obfuscators of both favors for some program families of interest. Third, we present an obfuscator for programs that test for membership in a hyperplane. This generalizes prior works that obfuscate equality testing. We prove the security of the obfuscator under a new strong variant of the Decisional Diffie-Hellman assumption that holds in the generic group model. Additionally, we show a cryptographic application of the new obfuscator to leak-ageresilient one-time digital signatures. The thesis also includes a survey of the prior results in the field.by Mayank Varia.Ph.D
Air quality of a steel city, rourkela, orissa
Full text linkThe iron and steel industry is an important basic industry because of high demand of iron and steel utilization, and the production process necessarily produces a significant volume of dust as well as gaseous effluents. An air quality monitoring program has been designed to collect the gaseous and particulate air pollutants from two sites (in front of Indira Gandhi Park & Udit Nagar) of the steel city, Rourkela. The PM10, TSP and gaseous air pollutants sulfur dioxide (SO2), nitrogen dioxide (NO2),ammonia (NH3) &carbon monoxide (CO) will be measured by high volume sampler .The average SO2,NO2,NH3,CO,PM10 and TSP concentration at Uditnagar (Residential site) during weekdays are 0.0155ppm, 0.3396ppm, 0.5119ppm, 0.70ppm, 24.50 µg/m3 and 158.38µg/m3 respectively. The average SO2, NO2, NH3, PM10 & TSP concentration at Uditnagar during weekends are 0.0179ppm, 0.2302ppm, 0.3238ppm, 0.75ppm, 8.57µg/m3 and 159.49µg/m3 respectively. The average SO2, NO2, NH3, PM10 & TSP concentration at Indira Gandhi Park (Industrial site) during weekdays are 0.0198pm, 0.2701ppm, 0.6744ppm, 0.54ppm, 17.76µg/m3 and 169.41µg/m3 respectively. The average SO2, NO2, NH3, PM10 & TSP concentration at Indra Gandhi Park weekends are 0.0193ppm, 0.2807ppm, 0.7847ppm, 0.24ppm, 22.81µg/m3 and 156.46µg/m3, respectively
A comparison of the triangle algorithm and sequential minimal optimization algorithm for solving the hard margin problem
No full textIn this article we consider the problem of testing, for two nite sets of points in the Euclidean space, if their convex hulls are disjoint and computing an optimal supporting hyperplane if so. This is a fundamental problem of classi cation in machine learning known as the hard-margin SVM. The problem can be formulated as a quadratic programming problem. The SMO algorithm [1] is the current state of art algorithm for solving it, but it does not answer the question of separability. An alternative to solving both problems is the triangle algorithm [2], a geometrically inspired algorithm, initially described for the convex hull membership problem [3], a fundamental problem in linear programming. First, we describe the experimental performance of the triangle algorithm for testing the intersection of two convex hulls. Next, we compare the performance of triangle algorithm with SMO for nding the optimal supporting hyperplane. Based on experimental results ranging up to 5000 points in each set in dimensions up to 10000, the triangle algorithm outperforms SMO.M.S.Includes bibliographical referencesby Mayank Gupt
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}
- …
