Dissertation and Thesis

Permanent URI for this communityhttps://dspace.isical.ac.in:4000/handle/10263/2146

Browse

Now showing 1 - 20 of 1341

2.5D Dual-Encoder U-Net for Lesion Segmentation in Chest CT Scans
(Indian Statistical Institute, Kolkata, 2025-06) Mukkara, Jagannath
Accurate segmentation of lesions in chest CT scans plays a vital role in diagnosing and monitoring pulmonary diseases such as COVID-19. In this, we introduce a novel 2.5D[1] dual-encoder U-Net model[2] that utilizes both the central slice and its neighboring slices to improve segmentation accuracy while keeping computational demands manageable. Our model incorporates residual connections[3] and feature fusion[4] to effectively merge multi-slice contextual information, overcoming the limitations found in traditional 2D and 3D methods. To ensure a reliable evaluation and avoid data leakage, we used patient-level data splitting. We validate our approach on a carefully curated chest CT dataset, showing enhanced segmentation performance and better generalization compared to standard U-Net models. Through extensive experiments, including ablation studies and visualizations, we demonstrate the advantages of combining 2.5D learning with a dual-encoder architecture for medical image segmentation tasks.
A Regression Tree Framework for Denoising and Monitoring of Image Data
(Indian Statistical Institute, 2026-07-02) Basak, Subhasish
The proliferation of advanced image acquisition technologies has led to the routine collection of large-scale image data across numerous scientific domains. This widespread reliance on image data accentuates the imperative to develop robust and efficient imaging techniques, which are essential for supporting modern applications across various scientific and industrial domains. This dissertation focuses on the development and analysis of methods for image denoising and image monitoring, two fundamental tasks in modern image analysis. A wide array of image denoising techniques exists in the literature, each tailored to handle specific types of noise or structural characteristics. However, no single method proves universally optimal, as each comes with its own advantages and trade-offs. In the first part of the dissertation, different configurations of local neighbourhoods are investigated, and an adaptive framework is proposed that combines these with local clustering-based smoothing to effectively harness the advantages of both methodologies. The dissertation then introduces a regression tree-based framework utilizing Oblique-axis Regression Trees (ORT) to estimate discontinuous regression functions in finite-dimensional spaces and applies this methodology to achieve effective image denoising. Due to an alternative set of assumptions on the underlying regression function, the overall structure of the proofs is substantially simpler than those typically found in the existing literature on regression trees. Finally, leveraging the ORT framework, the dissertation introduces an original approach to monitor drift patterns within an image sequence. Even though gradual temporal variations, known as drifts, are frequently observed in image sequences, drift monitoring remains an underexplored research area. This dissertation thus makes an effort to address that gap. Theoretical analysis and numerical studies, conducted on both simulated and real-world data, demonstrate the broad applicability and effectiveness of the proposed methods.
A Study of Prompt Tuning on Small Language Models(SLMs): A Controlled Benchmark and a Lightweight Instance-Aware Method
(2026-06-16) Sahith, Narkadamilli
Parameter-efficient fine-tuning (PEFT) adapts a frozen pre-trained language model by training only a small number of additional parameters. Among PEFT approaches, prompt tuning prepends trainable continuous vectors (soft prompts) to the input. A recurring finding in the literature is that prompt tuning is strongly scale dependent: it rivals full fine-tuning on very large models but lags on smaller ones. This dissertation studies prompt tuning specifically in the small-language-model (SLM) regime. We (i) re-implement a representative set of prompt-tuning methods—Prompt Tuning, P-Tuning v2, LoPT, DPT, DePT, ACCEPT, Residual Prompt Tuning, and PARA—within a single controlled harness, enabling a fair head-to-head comparison against full fine-tuning; (ii) propose IA-DePT, a lightweight instance-aware extension of Decomposed Prompt Tuning that conditions the short soft prompt on each input through a small, zero-initialised gate; and (iii) extend the benchmark beyond a single backbone and task, evaluating the full method suite on six backbone/task settings that span encoder–decoder (t5-small), encoder-only (BERT-base, RoBERTa-base, ELECTRA-small), and decoder-only (DistilGPT-2) architectures across the GLUE/SuperGLUE tasks RTE, WSC, CB, COPA, WiC, and MRPC. On RTE with t5-small, IA-DePT is the strongest parameter-efficient method in our benchmark (55.6% single-seed accuracy) and improves over its own base, DePT, by 6.5 points (53.6% vs. 47.1%, mean over three seeds) while adding only ≈16.9k parameters—a total trainable footprint of 0.05% of the backbone. Because the gate degrades exactly to DePT at initialisation, the comparison is a clean single-variable ablation. The cross-architecture study shows that the instance gate improves on DePT in five of the six settings on each setting’s primary metric (it ties or marginally regresses only on WiC, where every PEFT method sits at chance), so the benefit is broad but not universal. Our analysis characterises the accuracy/parameter trade-offs across method families, the strong effect of task difficulty on the small-model regime, and the role of instance-conditioning, including an honest discussion of why many prompt-tuning methods remain close to the chance baseline at this scale.
A Study of the SHA-2 Cryptographic Hash Family
(Indian Statistical Institute, Kolkata, 2009-02-01) Sanadhya Somitra Kumar
A Switch-Point-Aware Contrastive Approach to Sentiment Analysis of Hinglish Code-Mixed Text
(Indian Statistical Institute, 2026-06-15) Sahoo, Prasant Kumar
With the increasing use of social media in non-English-speaking regions, especially in India, people often use Romanized Hindi and English together in their online communication. In a single sentence, they frequently mix Romanized Hindi and English, creating code-mixed text. However, most multilingual transformer models are pre-trained primarily on monolingual data. As a result, NLP systems face challenges when processing code-mixed text, as a single word may be fragmented into meaningless subword pieces, making it difficult for the model to capture its semantic meaning accurately. In this dissertation, we propose a parameter efficient neural architecture consisting of three main components to address these challenges: First, there is a character-level CNN encoder, which handles spelling differences such as "nahi", "nahin", "nah", and "nai" through the chracter n-gram pattern. Next, there is a frozen XLM-R backbone(Conneau et al., 2019) , the top three layers, which are partly fine-tuned at a slower rate by which it provides rich cross lingual embeddings. Finally, there is a switch-point-aware bilingual gate that spots where the language label switches and blends two adapters using a learned gate weight.During training, it uses Supervised Contrastive Loss to learn better feature representations and Cross-Entropy Loss for classification. Since human annotators agreed on labels only 55% of the time, we use label smoothing to reflect this uncertainty and prevent the model from becoming overly confident in noisy labels. Evaluated on the SentiMix 2020 benchmark(Patwa et al., 2020), our proposed architecture achieves a weighted F1 score of 0.705, which outperforms the baseline model M-BERT (0.654 F1) and is comparable to fully fine-tuned transformer models while requiring only one-tenth of the trainable parameters.Adapter gate visualizations provide interpretable evidence that the gating mechanism captures linguistically meaningful codemixing structure. The architecture is designed to generalize to other code-mixed language pairs through its modular adapter design.
A1-homotopy types of A2 and A2 \ {(0, 0)}
(Indian Statistical Institute, Kolkata, 2024-12) Roy, Biman
Morel-Voevodsky developed A^1-homotopy theory which is a bridge between algebraic geometry and algebraic topology. In this thesis we study the A^1-connected component of a smooth variety in great detail. We have shown that the A^1-connected component of a smooth variety contains the information about the existence of affine lines in the variety. Using this and Miyanishi-Sugie's algebraic characterisation, we determine that the affine plane is the only A^1-contractible smooth affine surface over the field of characteristic zero. In the other part of the thesis, we studied the A^1-homotopy type of A^2-{(0,0)}. We showed that over the field of characteristic zero, if an open subvariety of a smooth affine surface is A^1-weakly equivalent to A^2-{(0,0)}, then it is isomorphic to A^2-{(0,0)}.
ABO blood-group gene frequencies in the Indian sub-continent: a statistical study of patterns of variation
(Indian Statistical Institute, Kolkata, 1980) Majumder, Partha P
Access structures for an image database
(Indian Statistical Institute, Kolkata, 1992) Kuila, Sudhansu Sekhar
Acyclicity Tests in Classes of Dense Digraphs in Streaming Model
(Indian Statistical Institute, Kolkata, 2020-07) Kundu, Madhumita
Graph is a popular model to represent highly structured data which involves entities who have pairwise relations between them. In many applications, computing graph theoretic properties after modelling the entire dataset as graph, provides us interesting informations which gives us insights about the whole dataset. However, in case of application, the datasets in question can be so large that it's di cult to store in the main memory and the dataset can even be dynamic(can change with time). These days in so many applications, the algorithm that requires to solve the problem which takes massive dataset as input, has limitations on time as well as space taken to store the information. These constraints leads us for the development of new techniques. Streaming model of computation takes all these challenges into account and provides us solutions with limited resources in cost of accuracy. Graph stream is a sequence of imcoming edges and we are only allowed to insert(insertion only model) or both insert and delete(dynamic model) into an initially empty graph. Finally our objective is to nd out certain properties of the graph at the end of the stream which minimizes the amount of space the algorithm uses. Sometimes this algorithm needs to provide the trade of between the space usage and the time taken. There is a large volume work on undirected graphs in streaming model but the area of directed graph stream is a pretty unexplored. In this project, we study the problem of testing acyclicity in dense digraphs in semi-streaming model. Here the graph on n vertices is presented as a stream of edges and using O(n polylog(n))-space, we must determine if it is acyclic or not
Adaptation-Based Classi ers for Handling Some Problems with Multi-Label Data
(Indian Statistical Institute, Kolkata, 2022-06) Law, Anwesha
The concept of multi-label (ML) data generalizes the association of instances to classes by labelling each data sample with more than one class simultaneously. Since this data can belong to more than one class at the same time, instances that are multi-label in nature, should not be forcefully assigned a single label. It needs to be handled in its original form. However, various problems arise while dealing with multi-label data. In this thesis, four such issues have been highlighted and dealt with. The first problem is the large input dimension that sometimes occurs in multi-label data. Dimensionality reduction of the features helps to strike a balance between the feature size, the number of samples and the output dimension. The next limitation is that of a complex decision space with overlapping class boundaries. This occurs due to the instances belonging to multiple classes simultaneously. Various approaches such as improving the feature to class mapping, increasing the class separability and simplifying the decision space have been implemented. The third drawback arises due to a large number of classes and label-sets in multi-label data, most of which are under-represented. This emphasizes the problem of class imbalance that widely prevails in multi-label data. This imbalance has been handled through the usage of customized classifiers suitable for the data at hand. Finally, the problem of class correlation is to be handled in this thesis. Multiple classes simultaneously assigned to every instance indicates a possibility of a few classes co-occurring on numerous occasions. These frequently co-occurring classes might have some correlation among them which have been identified and utilized to improve the multi-label classification performance.This thesis addresses the above-mentioned issues to perform efficient multi-label classification. Smaller components that target the individual issues have been incorporated to build large classification models. The first work aims to reduce feature dimensions and learn a better feature to class mapping for the complex decision space. A shallow but fast network known as extreme learning machines (ELMs) has been cascaded with autoencoders (AEs) to propose a network that can handle both issues. Two variations of the network have been proposed. To further explore the overlapping boundaries of ML data, the second contribution increases the separability of the complex decision space and also incorporates dimensionality reduction. Functional link artificial neural network (FLANN) has been adopted here for the unique functional expansion capability that transforms the features to a higher dimension thus making it considerably more separable. After identifying the best configuration of the network, it has then been integrated with autoencoders to reduce the functionally expanded feature dimension and bring additional transformation into the multi-label data. While these classifiers display improved performance, they do not consider the problems of class imbalance or label correlation. Hence, the third work builds a tree of classifiers that handles the problem of class imbalance, simplifies decision space for the ease of learning and preserves label correlations. A novel label-set proximity-based technique has been devised that simplifies boundaries and splits the data while preserving label correlations. Every split is learned by a classifier suited for the balanced or imbalanced data at hand. While handling multiple issues together successfully, this classifier tree model preserves label correlations but does not explicitly use them to improve classification performance. In this regard, the final contribution specifically extracts underlying label correlations from the data and associates them with predictions of existing multi-label classifiers to improve the overall performance. A novel frequent label-set mining technique generates rules that help to improve scores predicted by the existing multi-label algorithms. This thesis incorporates various elements to handle the problems of multi-label data and converges them to create cohesive models for multi-label classification.
Adaptive Spectral Trust Gate for Physics- Constrained Operator Learning
(2026-06-16) Chakraborty, Soham
Physics-informed machine learning improves the plausibility, data-efficiency and generalization of surrogate models by injecting prior physical knowledge into the learning process. The current approaches can be broadly divided into two main categories: soft constraints, which add a physics residual to the training loss but guarantee nothing at inference time, and hard constraints, which project the model output onto the constraint set exactly but apply the projection uniformly to every part of the signal — including parts that are dominated by noise, discretization error, or model mismatch, where the idealized physics is not actually trustworthy. This dissertation proposes the Adaptive Spectral Trust Gate (ASPINO), a mechanism that learns where to trust the physics. Operating in the Fourier domain on top of any surrogate model, a small gating network forms a per-mode convex combination of a data-driven soft path and a physics hard path. The gate is driven by features of the spectral coordinate and the spectral amplitude, so that it can apply the hard constraint in well-conditioned spectral regions and defer to the data-driven operator in regions corrupted by noise or aliasing. A single gate serves two very different hard paths - the linear Leray projection (incompressible flow) and a nonlinear rank-r SVD projection (massive-MIMO channel estimation). On the theoretical side, we give an empirical-Rademacher-complexity analysis: an unconditional safety floor — the gated class never exceeds the soft path it wraps — and, under a stated low-rank-transfer assumption, a capacity-reduction factor of 1 − ¯α (1 − √ρr), where ¯α is the fraction of capacity routed through the hard path. On Kolmogorov-flow denoising ASPINO is simultaneously the most accurate and near physical, dominating the unconstrained, hard and soft baselines; on ray-traced MIMO it improves a strong physics-informed baseline across all pilot budgets and SNRs without ever regressing. A third study, zero-shot super-resolution on the Poisson equation, confirms the discretization invariance of the gated construction. ASPINO is discretization-invariant and “plug-and-play” over the underlying operator.
Addressing class imbalance problems to improve animal detection through aerial image data
(Indian Statistical Institute, Kolkata, 2025-06) Koushal, Suryang
Monitoring animal populations in wildlife reserves is essential for conservation, especially for endangered species, but manual censuses are costly, risky, and logistically challenging due to vast, inaccessible terrains. Unmanned Aerial Vehicles (UAVs) with digital cameras provide a safer, scalable solution for collecting aerial imagery to estimate animal populations. However, semi-automated processing of these images faces significant challenges due to class imbalance in datasets, including foreground-background disparities, where background terrain dominates over sparse animal instances, and inter-class imbalances from uneven species representation and varied visual appearances (e.g., species, sizes, fur patterns) against diverse backgrounds like deserts or forests. These imbalances hinder Convolutional Neural Networks (CNNs) used for object detection, leading to inaccurate population estimates. This project addresses these issues using a dataset of 561 aerial images from Tsavo National Parks (March 2014) and Laikipia-Samburu Ecosystem (May 2015), collected by the Kenya Wildlife Service. We propose a clustering-based approach to categorize background terrain into distinct classes (e.g., desert, grassland), aiming to mitigate imbalances and improve animal detection accuracy in UAV imagery, supporting reliable, data-driven conservation strategies.
Administrative document processing
(Indian Statistical Institute, Kolkata, 2016) Chandra, Satish
Advanced Techniques in Symmetric Key Cryptanalysis
(Indian Statistical Institute, Kolkata, 2024-07) Chakraborty, Debasmita
Symmetric key cryptographic primitives are essential tools used extensively in daily digital interactions. These primitives are mainly designed to provide three key services: ensuring data confidentiality, maintaining data integrity, and verifying the authenticity of data sources. The primary types of symmetric key primitives that deliver these services include block ciphers, stream ciphers, hash functions, message authentication codes, and authenticated encryption with associated data. This thesis mainly explores the security analysis of hash functions, several block ciphers, and stream ciphers using some advanced cryptanalytic techniques. We begin by examining the collision security of a hash function, specifically under the assumption that the underlying compression functions are collision-resistant. This characteristic is termed the collision-resistance preserving property of a hash function. Notably, both the Merkle-Damgård and Merkle tree hash structures exhibit this property, prompting the question of whether it is possible to reduce the number of underlying compression function calls while maintaining the collision-resistance preserving property. In pursuit of this question, we prove that for an ℓn-to-sn-bit collision-preserving hash function, designed using r tn-to-n-bit compression function calls, it must hold that r ≥ ⌈(ℓ − s)/(t − 1)⌉, assuming all operations other than the compression function are linear. Shifting our focus, we delve into advanced techniques for enhanced cryptanalysis of block and stream ciphers. Initially, we concentrate on the impossible differential (ID) and zero correlation (ZC) attacks, which are pivotal cryptanalytic methods for block ciphers. We introduce an advanced, unified constraint programming (CP) approach based on satisfiability for identifying ID distinguishers in ARX and AndRX ciphers alongside a similar method for identifying ZC distinguishers. Furthermore, we extend our novel model to formulate a unified optimization problem that incorporates the distinguisher and key recovery for AndRX designs. Our approach not only enhances ID attacks but also unveils new distinguishers for various ciphers, including SIMON, SPECK, Simeck, ChaCha, Chaskey, LEA, and SipHash. Another significant cryptanalytic technique, particularly applicable to the analysis of block and stream ciphers, is the division property—an advanced version of integral cryptanalysis. Here, we explore the feasibility of the MILP method for the bit-based division property using three subsets (BDPT) propagation in ciphers with complex linear layers. We apply our novel method to discover integral distinguishers based on BDPT for the SIMON, SIMON(102), PRINCE, MANTIS, PRIDE, and KLEIN block ciphers. The integral distinguishers identified by our method are superior to or consistent with the longest existing distinguishers. Finally, we investigate the cube attack, a powerful cryptanalytic technique against stream ciphers. We study the NIST lightweight 3rd round candidate Grain-128AEAD through the lens of division property-based cube attacks. Initially, we introduce some effective cubes and construct an algorithm to identify conditional key bits for these cubes in Grain-128AEAD. Subsequently, we employ the three-subset division property without unknown subsets based cube attacks to recover exact superpolies for Grain-128AEAD in the weak-key setting, yielding improved results.
Adversarial Attack on Neural Machine Translation System
(Indian Statistical Institute, Kolkata, 2019-06) Abijith, K P
Nowadays Deep Neural Network based solutions are deployed to solve numerous tasks. Thus, it has become absolutely important to study the robustness of these systems. Machine Translation is one of the popular applications of Deep Neural Networks. This thesis studies the robustness of Neural Machine Translation systems by generating adversarial examples with the objective to fool the model. Whenever there is a change in the source, i.e. when a word in the input sentence is replaced by an unrelated word, the translation system is supposed to re ect the changes while doing translation. These unwanted invariance learned by the model is undesirable. With intention to exploit this undesirable property learned by a Neural Machine Translation system we design an attack called: Invariance-based targeted attack. This attack introduces multiple changes(replacement of words) to the original input sentence, keeping the translation unchanged. In-order to facilitate the explanation of the design of the attack we introduce two methods: (i) Min-Grad method: To identify the position where a replacement of the word makes the least change in the translation, and (ii) Soft-Attn method: To search for a new word to replace, given a list of choices. The initial part of the report explain the preliminary explorations we did in-order to get some insights on how to do the problem formulation. These experiments are run on LSTM based models with single replacement policy. Using the learning from the rst part we extend the experiments to Transformer and BLSTM based models, which are considered as the state-of-the-art systems for machine translation.
Agricultural tenancy in palanpur
(Indian Statistical Institute,Delhi, 1992) Sharma, Naresh Kumar
Agriculture trade and protectionism
(Indian Statistical Institute, Kolkata, 2013-07) Basu, Debasmita
Air-Writing Recognition
(Indian Statistical Institute, 2026-06-15) Shukla, Gaurang
Air-writing is the act of tracing characters or words in free space with a fingertip, recorded by a camera, giving a touch-free input modality for smart displays, augmented and virtual reality, and assistive interfaces. It is difficult because the finger never lifts: connecting strokes join adjacent letters with no pen-up signal to mark boundaries, and the same word varies widely in scale, position, and slant across writers. The WiTA benchmark of Kim et al. provides a large, person-disjoint dataset and a baseline that treats each clip as RGB video, recognised by a spatio-temporal 3D residual network trained with a CTC objective, reaching a character error rate (CER) of 0.292 on the English subset. The main goal of this dissertation was to improve on this error rate, which we achieve: we replace raw video with an explicit fingertip-trajectory sequence extracted from hand landmarks, fed to a Conformer encoder with a joint CTC/attention head. The resulting system attains a test CER of 0.219, improving on the published 0.292 of Kim et al. and 0.299 of Tan et al. by 15–27% relative.
Alfsen-errors structure topology in the theory of complex L1-preduals
(Indian Statistical Institute, Kolkata, 1981) Rao, T S S R K
Algorithm for mapping boolean network to LUT based FPGAs
(Indian Statistical Institute, Kolkata, 2001) Bhattacharyya, Jayasri