Searcharxiv⌕ Search

arXiv subjects

Subhashis Banerjee

Publications and source records attributed to Subhashis Banerjee.

At least 19 recordsLinked to original sources
arXiv2026

Are You Thinking What I am Thinking? : Examining Conceptual Separation in Neural Architectures

Neural networks are increasingly employed to identify both well-defined and ambiguous concepts, yet output-level metrics reveal little about how those concepts are represented internally. Our study asks if these networks exhibit \textit{conceptual separation}: if examples of the same concept form coherent representations, and whether related concepts lie closer together in the representation space. We examine this conceptual organisation in Convolutional Neural Networks (CNNs) and Large Language Models (LLMs) through geometric and distributional analysis of their internal activations. In CNNs, familiar ImageNet concepts form coherent and semantically ordered representations, while this coherence weakens for unseen concepts and suffers within-class domain shift. In LLMs, clearly distinct domains remain well separated, related subdomains move closer together, and the distinction between ambiguous topics collapses at both the mean and covariance level. These results suggest that conceptual separation can reveal structure that output accuracy alone cannot, and may serve as a useful diagnostic of how robustly a model represents the concepts it is asked to identify. Code and data available on \href{https://github.com/JaeeRoshniCapstoneProject/Are-You-Thinking-What-I-m-Thinking-Examining-Conceptual-Separation-in-Neural-Architectures}{GitHub}.

cs.LG↗
arXiv2026

Validity, Reliability, and Transparency in Artificial Intelligence Regulation

Artificial intelligence (AI) systems increasingly mediate decisions affecting individuals and societies. Existing data protection frameworks address certain privacy-related harms, particularly those arising from data leakage, re-identification, and profiling. However, they inadequately capture a more fundamental risk: unreliable or unjustified inference produced by AI systems even when data collection and processing are legitimate. This article argues that modern AI raises distinct concerns of construct validity, confounding, representativeness, distribution shift, and fairness trade-offs that require specialised regulatory attention. In the context of AI, transparency and explainability acquire distinct and significantly more challenging meanings than in conventional software. A substantial body of work in critical data studies and the measurement-theoretic literature has diagnosed these epistemological limitations. This article's contribution is to derive from that diagnosis a structured and operationalizable regulatory framework. We argue that validity of inference should function as a precondition for proportionality assessment and deployment approval --- a move that existing frameworks, including the EU AI Act's domain-based risk tiers, do not make. We ground this argument in the constitutional principle of informational self-determination articulated in the Indian Supreme Court's \emph{Puttaswamy} judgement, extending its reach from data collection to the legitimacy of use of data. Effective governance must therefore incorporate AI-specific validity assessment, post-deployment monitoring, and proportionality assessments grounded in structured articulation of both epistemic risk and potential benefit.

cs.CY↗
arXiv2026

A New Framework to Analyse the Distributional Robustness of Deep Neural Networks

Deep neural networks have achieved impressive performance on a variety of tasks, but their brittleness to distributional shifts remains a significant barrier to real-world deployment. In this paper, we propose a framework to analyse and quantify the distributional robustness of neural networks by studying the interactions between layer weights and activations. We model these interactions using Bernoulli distributions, using the separation between classes as a diagnostic proxy for robustness. We demonstrate the usefulness of this framework through models trained on CIFAR-10 and ImageNet. We show that our proposed metrics can distinguish between networks that have memorised their training data and those that have not. We also perform analogous experiments in the activation space and find that the same properties do not hold up. Additionally, we investigate the behaviour of our metrics under various distribution shifts and show that these shifts reduce separation under our path-based diagnostics. Our results suggest that this framework provides useful model-level diagnostics of representation structure and robustness.

cs.LG↗
arXiv2024

Publicly auditable privacy-preserving electoral rolls

While existing literature on electronic voting has extensively addressed verifiability of voting protocols, the vulnerability of electoral rolls in large public elections remains a critical concern. To ensure integrity of electoral rolls, the current practice is to either make electoral rolls public or share them with the political parties. However, this enables construction of detailed voter profiles and selective targeting and manipulation of voters, thereby undermining the fundamental principle of free and fair elections. In this paper, we study the problem of designing publicly auditable yet privacy-preserving electoral rolls. We first formulate a threat model and provide formal security definitions. We then present a protocol for creation, maintenance and usage of electoral rolls that mitigates the threats. Eligible voters can verify their inclusion, whereas political parties and auditors can statistically audit the electoral roll. Further, the audit can also detect polling-day ballot stuffing and denials to eligible voters by malicious polling officers. The entire electoral roll is never revealed, which prevents any large-scale systematic voter targeting and manipulation.

cs.CR↗
arXiv2023

Traceable mixnets

We introduce the notion of \emph{traceable mixnets}. In a traditional mixnet, multiple mix-servers jointly permute and decrypt a list of ciphertexts to produce a list of plaintexts, along with a proof of correctness, such that the association between individual ciphertexts and plaintexts remains completely hidden. However, in many applications, the privacy-utility tradeoff requires answering some specific queries about this association, without revealing any information beyond the query result. We consider queries of the following types: a) given a ciphertext in the mixnet input list, whether it encrypts one of a given subset of plaintexts in the output list, and b) given a plaintext in the mixnet output list, whether it is a decryption of one of a given subset of ciphertexts in the input list. Traceable mixnets allow the mix-servers to jointly prove answers to the above queries to a querier such that neither the querier nor a threshold number of mix-servers learn any information beyond the query result. Further, if the querier is not corrupted, the corrupted mix-servers do not even learn the query result. We first comprehensively formalise these security properties of traceable mixnets and then propose a construction of traceable mixnets using novel distributed zero-knowledge proofs (ZKPs) of set membership and of a statement we call reverse set membership. Although set membership has been studied in the single-prover setting, the main challenge in our distributed setting lies in making sure that none of the mix-servers learn the association between ciphertexts and plaintexts during the proof. We implement our distributed ZKPs and show that they are faster than state-of-the-art by at least one order of magnitude.

cs.CR↗
arXiv2022

From Forks to Forceps: A New Framework for Instance Segmentation of Surgical Instruments

Minimally invasive surgeries and related applications demand surgical tool classification and segmentation at the instance level. Surgical tools are similar in appearance and are long, thin, and handled at an angle. The fine-tuning of state-of-the-art (SOTA) instance segmentation models trained on natural images for instrument segmentation has difficulty discriminating instrument classes. Our research demonstrates that while the bounding box and segmentation mask are often accurate, the classification head mis-classifies the class label of the surgical instrument. We present a new neural network framework that adds a classification module as a new stage to existing instance segmentation models. This module specializes in improving the classification of instrument masks generated by the existing model. The module comprises multi-scale mask attention, which attends to the instrument region and masks the distracting background features. We propose training our classifier module using metric learning with arc loss to handle low inter-class variance of surgical instruments. We conduct exhaustive experiments on the benchmark datasets EndoVis2017 and EndoVis2018. We demonstrate that our method outperforms all (more than 18) SOTA methods compared with, and improves the SOTA performance by at least 12 points (20%) on the EndoVis2017 benchmark challenge and generalizes effectively across the datasets.

cs.CV↗
arXiv2022

Ballot stuffing and participation privacy in pollsite voting

We study the problem of simultaneously addressing both ballot stuffing and participation privacy for pollsite voting systems. Ballot stuffing is the attack where fake ballots (not cast by any eligible voter) are inserted into the system. Participation privacy is about hiding which eligible voters have actually cast their vote. So far, the combination of ballot stuffing and participation privacy has been mostly studied for internet voting, where voters are assumed to own trusted computing devices. Such approaches are inapplicable to pollsite voting where voters typically vote bare handed. We present an eligibility audit protocol to detect ballot stuffing in pollsite voting protocols. This is done while protecting participation privacy from a remote observer - one who does not physically observe voters during voting. Our protocol can be instantiated as an additional layer on top of most existing pollsite E2E-V voting protocols. To achieve our guarantees, we develop an efficient zero-knowledge proof (ZKP), that, given a value $v$ and a set $\Phi$ of commitments, proves $v$ is committed by some commitment in $\Phi$, without revealing which one. We call this a ZKP of reverse set membership because of its relationship to the popular ZKPs of set membership. This ZKP may be of independent interest.

cs.CR↗
arXiv2021

QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation - Analysis of Ranking Scores and Benchmarking Results

Deep learning (DL) models have provided state-of-the-art performance in various medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g., tumor and lesion sub-regions) is particularly challenging, and potential errors hinder translating DL models into clinical workflows. Quantifying the reliability of DL model predictions in the form of uncertainties could enable clinical review of the most uncertain regions, thereby building trust and paving the way toward clinical translation. Several uncertainty estimation methods have recently been introduced for DL medical image segmentation tasks. Developing scores to evaluate and compare the performance of uncertainty measures will assist the end-user in making more informed decisions. In this study, we explore and evaluate a score developed during the BraTS 2019 and BraTS 2020 task on uncertainty quantification (QU-BraTS) and designed to assess and rank uncertainty estimates for brain tumor multi-compartment segmentation. This score (1) rewards uncertainty estimates that produce high confidence in correct assertions and those that assign low confidence levels at incorrect assertions, and (2) penalizes uncertainty measures that lead to a higher percentage of under-confident correct assertions. We further benchmark the segmentation uncertainties generated by 14 independent participating teams of QU-BraTS 2020, all of which also participated in the main BraTS segmentation task. Overall, our findings confirm the importance and complementary value that uncertainty estimates provide to segmentation algorithms, highlighting the need for uncertainty quantification in medical image analyses. Finally, in favor of transparency and reproducibility, our evaluation code is made publicly available at: https://github.com/RagMeh11/QU-BraTS.

eess.IV↗
arXiv2021

Analysis of MRI Biomarkers for Brain Cancer Survival Prediction

Prediction of Overall Survival (OS) of brain cancer patients from multi-modal MRI is a challenging field of research. Most of the existing literature on survival prediction is based on Radiomic features, which does not consider either non-biological factors or the functional neurological status of the patient(s). Besides, the selection of an appropriate cut-off for survival and the presence of censored data create further problems. Application of deep learning models for OS prediction is also limited due to the lack of large annotated publicly available datasets. In this scenario we analyse the potential of two novel neuroimaging feature families, extracted from brain parcellation atlases and spatial habitats, along with classical radiomic and geometric features; to study their combined predictive power for analysing overall survival. A cross validation strategy with grid search is proposed to simultaneously select and evaluate the most predictive feature subset based on its predictive power. A Cox Proportional Hazard (CoxPH) model is employed for univariate feature selection, followed by the prediction of patient-specific survival functions by three multivariate parsimonious models viz. Coxnet, Random survival forests (RSF) and Survival SVM (SSVM). The brain cancer MRI data used for this research was taken from two open-access collections TCGA-GBM and TCGA-LGG available from The Cancer Imaging Archive (TCIA). Corresponding survival data for each patient was downloaded from The Cancer Genome Atlas (TCGA). A high cross validation $C-index$ score of $0.82\pm.10$ was achieved using RSF with the best $24$ selected features. Age was found to be the most important biological predictor. There were $9$, $6$, $6$ and $2$ features selected from the parcellation, habitat, radiomic and region-based feature groups respectively.

q-bio.QM↗
arXiv2020

REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions

Deep Neural Networks (DNNs) are often criticized for being susceptible to adversarial attacks. Most successful defense strategies adopt adversarial training or random input transformations that typically require retraining or fine-tuning the model to achieve reasonable performance. In this work, our investigations of intermediate representations of a pre-trained DNN lead to an interesting discovery pointing to intrinsic robustness to adversarial attacks. We find that we can learn a generative classifier by statistically characterizing the neural response of an intermediate layer to clean training samples. The predictions of multiple such intermediate-layer based classifiers, when aggregated, show unexpected robustness to adversarial attacks. Specifically, we devise an ensemble of these generative classifiers that rank-aggregates their predictions via a Borda count-based consensus. Our proposed approach uses a subset of the clean training data and a pre-trained model, and yet is agnostic to network architectures or the adversarial attack generation method. We show extensive experiments to establish that our defense strategy achieves state-of-the-art performance on the ImageNet validation set.

cs.CV↗
arXiv2020

An operational architecture for privacy-by-design in public service applications

Governments around the world are trying to build large data registries for effective delivery of a variety of public services. However, these efforts are often undermined due to serious concerns over privacy risks associated with collection and processing of personally identifiable information. While a rich set of special-purpose privacy-preserving techniques exist in computer science, they are unable to provide end-to-end protection in alignment with legal principles in the absence of an overarching operational architecture to ensure purpose limitation and protection against insider attacks. This either leads to weak privacy protection in large designs, or adoption of overly defensive strategies to protect privacy by compromising on utility. In this paper, we present an operational architecture for privacy-by-design based on independent regulatory oversight stipulated by most data protection regimes, regulated access control, purpose limitation and data minimisation. We briefly discuss the feasibility of implementing our architecture based on existing techniques. We also present some sample case studies of privacy-preserving design sketches of challenging public service applications.

cs.CR↗
arXiv2020

CovidAID: COVID-19 Detection Using Chest X-Ray

The exponential increase in COVID-19 patients is overwhelming healthcare systems across the world. With limited testing kits, it is impossible for every patient with respiratory illness to be tested using conventional techniques (RT-PCR). The tests also have long turn-around time, and limited sensitivity. Detecting possible COVID-19 infections on Chest X-Ray may help quarantine high risk patients while test results are awaited. X-Ray machines are already available in most healthcare systems, and with most modern X-Ray systems already digitized, there is no transportation time involved for the samples either. In this work we propose the use of chest X-Ray to prioritize the selection of patients for further RT-PCR testing. This may be useful in an inpatient setting where the present systems are struggling to decide whether to keep the patient in the ward along with other patients or isolate them in COVID-19 areas. It would also help in identifying patients with high likelihood of COVID with a false negative RT-PCR who would need repeat testing. Further, we propose the use of modern AI techniques to detect the COVID-19 patients using X-Ray images in an automated manner, particularly in settings where radiologists are not available, and help make the proposed testing technology scalable. We present CovidAID: COVID-19 AI Detector, a novel deep neural network based model to triage patients for appropriate testing. On the publicly available covid-chestxray-dataset [2], our model gives 90.5% accuracy with 100% sensitivity (recall) for the COVID-19 infection. We significantly improve upon the results of Covid-Net [10] on the same dataset.

eess.IV↗
arXiv2019

OpenVoting: Recoverability from Failures in Dual Voting

In this paper we address the problem of recovery from failures without re-running entire elections when elections fail to verify. We consider the setting of \emph{dual voting} protocols, where the cryptographic guarantees of end-to-end verifiable voting (E2E-V) are combined with the simplicity of audit using voter-verified paper records (VVPR). We first consider the design requirements of such a system and then suggest a protocol called \emph{OpenVoting}, which identifies a verifiable subset of error-free votes consistent with the VVPRs, and the polling booths corresponding to the votes that fail to verify with possible reasons for the failures. To an ordinary voter \emph{OpenVoting} looks just like an old fashioned paper based voting system, with minimal additional cognitive overload.

cs.CR↗
arXiv2019

Deep Radiomics for Brain Tumor Detection and Classification from Multi-Sequence MRI

Glioma constitutes 80% of malignant primary brain tumors and is usually classified as HGG and LGG. The LGG tumors are less aggressive, with slower growth rate as compared to HGG, and are responsive to therapy. Tumor biopsy being challenging for brain tumor patients, noninvasive imaging techniques like Magnetic Resonance Imaging (MRI) have been extensively employed in diagnosing brain tumors. Therefore automated systems for the detection and prediction of the grade of tumors based on MRI data becomes necessary for assisting doctors in the framework of augmented intelligence. In this paper, we thoroughly investigate the power of Deep ConvNets for classification of brain tumors using multi-sequence MR images. We propose novel ConvNet models, which are trained from scratch, on MRI patches, slices, and multi-planar volumetric slices. The suitability of transfer learning for the task is next studied by applying two existing ConvNets models (VGGNet and ResNet) trained on ImageNet dataset, through fine-tuning of the last few layers. LOPO testing, and testing on the holdout dataset are used to evaluate the performance of the ConvNets. Results demonstrate that the proposed ConvNets achieve better accuracy in all cases where the model is trained on the multi-planar volumetric dataset. Unlike conventional models, it obtains a testing accuracy of 95% for the low/high grade glioma classification problem. A score of 97% is generated for classification of LGG with/without 1p/19q codeletion, without any additional effort towards extraction and selection of features. We study the properties of self-learned kernels/ filters in different layers, through visualization of the intermediate layer outputs. We also compare the results with that of state-of-the-art methods, demonstrating a maximum improvement of 7% on the grading performance of ConvNets and 9% on the prediction of 1p/19q codeletion status.

cs.CV↗
arXiv2018

Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge

Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, i.e., peritumoral edematous/invaded tissue, necrotic core, active and non-enhancing core. This intrinsic heterogeneity is also portrayed in their radio-phenotype, as their sub-regions are depicted by varying intensity profiles disseminated across multi-parametric magnetic resonance imaging (mpMRI) scans, reflecting varying biological properties. Their heterogeneous shape, extent, and location are some of the factors that make these tumors difficult to resect, and in some cases inoperable. The amount of resected tumor is a factor also considered in longitudinal scans, when evaluating the apparent tumor for potential diagnosis of progression. Furthermore, there is mounting evidence that accurate segmentation of the various tumor sub-regions can offer the basis for quantitative image analysis towards prediction of patient overall survival. This study assesses the state-of-the-art machine learning (ML) methods used for brain tumor image analysis in mpMRI scans, during the last seven instances of the International Brain Tumor Segmentation (BraTS) challenge, i.e., 2012-2018. Specifically, we focus on i) evaluating segmentations of the various glioma sub-regions in pre-operative mpMRI scans, ii) assessing potential tumor progression by virtue of longitudinal growth of tumor sub-regions, beyond use of the RECIST/RANO criteria, and iii) predicting the overall survival from pre-operative mpMRI scans of patients that underwent gross total resection. Finally, we investigate the challenge of identifying the best ML algorithms for each of these tasks, considering that apart from being diverse on each instance of the challenge, the multi-institutional mpMRI BraTS dataset has also been a continuously evolving/growing dataset.

cs.CV↗
arXiv2018

Authorisation and access control architecture as a framework for data and privacy protection

Privacy protection in digital databases does not demand that data should not be collected, stored or used, but that there should be guarantees that the data can only be used for pre-approved and legitimate purposes. We argue that a data protection law based on traditional understanding of privacy protection and detection of privacy infringements is unlikely to be successful, and that what is required is a law based on an understanding of the architectural requirements of authorisation, audit and access control in real-time. Despite the protection principles being sound, privacy protection in digital databases has been less than effective, anywhere, mainly because of weak enforcement methods.

cs.CY↗
arXiv2017

On monitoring development indicators using high resolution satellite images

We develop a machine learning based tool for accurate prediction of socio-economic indicators from daytime satellite imagery. The diverse set of indicators are often not intuitively related to observable features in satellite images, and are not even always well correlated with each other. Our predictive tool is more accurate than using night light as a proxy, and can be used to predict missing data, smooth out noise in surveys, monitor development progress of a region, and flag potential anomalies. Finally, we use predicted variables to do robustness analysis of a regression study of high rate of stunting in India.

econ.EM↗