arXiv subjects

Kapil Ahuja

Publications and source records attributed to Kapil Ahuja.

At least 19 recordsLinked to original sources

arXiv2026

An Improved Adaptive PID Optimizer with Enhanced Convergence and Stability for Deep Learning

Optimization is essential in deep learning. The foundational method upon which most optimizers are built is momentum-based stochastic gradient descent. However, it suffers from two key drawbacks. First, it has noisy and varying gradients, and second, it has an overshoot phenomenon. To address noisy gradients, Adam was proposed, which remains the most widely used adaptive optimizer. To address the overshoot phenomenon, a control-theory-based PID optimizer was proposed. To tackle both the limitations within a single framework, several variants of Adaptive PID (AdaPID) have recently been proposed. Although AdaPID performs well, it still inherits two critical drawbacks from Adam, namely convergence and stability issues. In this work, we address both these limitations. To fix the convergence issue, we uniquely integrate the idea of using a non-increasing effective learning rate into AdaPID (originally proposed in AMSGrad, an extension of Adam). To fix the stability issue, we innovatively integrate a gradient difference based modulation factor into AdaPID (originally proposed in DiffGrad, another extension of Adam). Combining both these ideas in AdaPID, results in our novel IAdaPID-ADG optimizer. We evaluate our proposed optimizer on multiple datasets, including benchmark datasets (MNIST and CIFAR10) and real-world datasets (IARC and AnnoCerv). The IAdaPID-ADG substantially outperforms all competing optimizers. Additionally, we perform an ablation study on the MNIST dataset to demonstrate the contribution of each added component.

cs.LG↗

arXiv2026

ResGene-T: A Tensor-Based Residual Network Approach for Genomic Prediction

In this work, we propose a new deep learning model for Genomic Prediction (GP), which involves correlating genotypic data with phenotypic. The genotypes are typically fed as a sequence of characters to the 1D-Convolution Neural Network layer of the underlying deep learning model. Inspired by earlier work that represented genotype as a 2D-image for genotype-phenotype classification, we extend this idea to GP, which is a regression task. We use a ResNet-18 as the underlying architecture, and term this model as ResGene-2D. Although the 2D-image representation captures biological interactions well, it requires all the layers of the model to do so. This limits training efficiency. Thus, as seen in the earlier work that proposed a 2D-image representation, our ResGene-2D performs almost the same as other models (3% improvement). To overcome this, we propose a novel idea of converting the 2D-image into a 3D/ tensor and feed this to the ResNet-18 architecture, and term this model as ResGene-T. We evaluate our proposed models on three crop species having ten phenotypic traits and compare it with seven most popular models (two statistical, two machine learning, and three deep learning). ResGene-T performs the best among all these seven methods (gains from 14.51% to 41.51%).

cs.LG↗

arXiv2025

Accurate Thyroid Cancer Classification using a Novel Binary Pattern Driven Local Discrete Cosine Transform Descriptor

In this study, we develop a new CAD system for accurate thyroid cancer classification with emphasis on feature extraction. Prior studies have shown that thyroid texture is important for segregating the thyroid ultrasound images into different classes. Based upon our experience with breast cancer classification, we first conjuncture that the Discrete Cosine Transform (DCT) is the best descriptor for capturing textural features. Thyroid ultrasound images are particularly challenging as the gland is surrounded by multiple complex anatomical structures leading to variations in tissue density. Hence, we second conjuncture the importance of localization and propose that the Local DCT (LDCT) descriptor captures the textural features best in this context. Another disadvantage of complex anatomy around the thyroid gland is scattering of ultrasound waves resulting in noisy and unclear textures. Hence, we third conjuncture that one image descriptor is not enough to fully capture the textural features and propose the integration of another popular texture capturing descriptor (Improved Local Binary Pattern, ILBP) with LDCT. ILBP is known to be noise resilient as well. We term our novel descriptor as Binary Pattern Driven Local Discrete Cosine Transform (BPD-LDCT). Final classification is carried out using a non-linear SVM. The proposed CAD system is evaluated on the only two publicly available thyroid cancer datasets, namely TDID and AUITD. The evaluation is conducted in two stages. In Stage I, thyroid nodules are categorized as benign or malignant. In Stage II, the malignant cases are further sub-classified into TI-RADS (4) and TI-RADS (5). For Stage I classification, our proposed model demonstrates exceptional performance of nearly 100% on TDID and 97% on AUITD. In Stage II classification, the proposed model again attains excellent classification of close to 100% on TDID and 99% on AUITD.

cs.CV↗

arXiv2025

Chameleon2++: An Efficient and Scalable Variant Of Chameleon Clustering

Hierarchical clustering remains a fundamental challenge in data mining, particularly when dealing with large-scale datasets where traditional approaches fail to scale effectively. Recent Chameleon-based algorithms - Chameleon2, M-Chameleon, and INNGS-Chameleon have proposed advanced strategies but they still suffer from $O(n^2)$ computational complexity, especially for large datasets. With Chameleon2 as the base algorithm, we introduce Chameleon2++ that addresses this challenge. Our algorithm has three parts. First, Graph Generation - we propose an approximate $k$-NN search instead of an exact one, specifically we integrate with the Annoy algorithm. This results in fast approximate nearest neighbor computation, significantly reducing the graph generation time. Second, Graph Partitioning - we propose use of a multi-level partitioning algorithm instead of a recursive bisection one. Specifically we adapt the hMETIS algorithm instead of the FM. This is because multi-level algorithms are robust to approximation introduced in the graph generation phase yielding higher-quality partitions, and that too with minimum configuration requirements. Third, Merging - we retain the flood fill heuristic that ensures connected balanced components in the partitions as well as efficient partition merging criteria leading to the final clusters. These enhancements reduce the overall time complexity to $O(n\log n)$, achieving scalability. On real-world benchmark datasets used in prior Chameleon works, Chameleon2++ delivers an average of 4% improvement in clustering quality. This demonstrates that algorithmic efficiency and clustering quality can co-exist in large-scale hierarchical clustering.

cs.LG↗

arXiv2024

AI Algorithm for Predicting and Optimizing Trajectory of UAV Swarm

This paper explores the application of Artificial Intelligence (AI) techniques for generating the trajectories of fleets of Unmanned Aerial Vehicles (UAVs). The two main challenges addressed include accurately predicting the paths of UAVs and efficiently avoiding collisions between them. Firstly, the paper systematically applies a diverse set of activation functions to a Feedforward Neural Network (FFNN) with a single hidden layer, which enhances the accuracy of the predicted path compared to previous work. Secondly, we introduce a novel activation function, AdaptoSwelliGauss, which is a sophisticated fusion of Swish and Elliott activations, seamlessly integrated with a scaled and shifted Gaussian component. Swish facilitates smooth transitions, Elliott captures abrupt trajectory changes, and the scaled and shifted Gaussian enhances robustness against noise. This dynamic combination is specifically designed to excel in capturing the complexities of UAV trajectory prediction. This new activation function gives substantially better accuracy than all existing activation functions. Thirdly, we propose a novel Integrated Collision Detection, Avoidance, and Batching (ICDAB) strategy that merges two complementary UAV collision avoidance techniques: changing UAV trajectories and altering their starting times, also referred to as batching. This integration helps overcome the disadvantages of both - reduction in the number of trajectory manipulations, which avoids overly convoluted paths in the first technique, and smaller batch sizes, which reduce overall takeoff time in the second.

cs.RO↗

arXiv2024

Block-Fused Attention-Driven Adaptively-Pooled ResNet Model for Improved Cervical Cancer Classification

Cervical cancer is the second most common cancer among women and a leading cause of mortality. Many attempts have been made to develop an effective Computer Aided Diagnosis (CAD) system; however, their performance remains limited. Using pretrained ResNet-50/101/152, we propose a novel CAD system that significantly outperforms prior approaches. Our novel model has three key components. First, we extract detailed features (color, edges, and texture) from early convolution blocks and the abstract features (shapes and objects) from later blocks, as both are equally important. This dual-level feature extraction is a new paradigm in cancer classification. Second, a non-parametric 3D attention module is uniquely embedded within each block for feature enhancement. Third, we design a theoretically motivated innovative adaptive pooling strategy for feature selection that applies Global Max Pooling to detailed features and Global Average Pooling to abstract features. These components form our Proposed Block-Fused Attention-Driven Adaptively-Pooled ResNet (BF-AD-AP-ResNet) model. To further strengthen learning, we introduce a Tri-Stream model, which unifies the enhanced features from three BF-AD-AP-ResNets. An SVM classifier is employed for final classification. We evaluate our models on two public datasets, IARC and AnnoCerv. On IARC, the base ResNets achieve an average performance of 90.91%, while our model achieves an excellent performance of 98.63%. On AnnoCerv, the base ResNets reach to 87.68%, and our model improves this significantly, reaching 93.39%. Our approach outperforms the best existing method on IARC by an average of 14.55%. For AnnoCerv, no prior competitive works are available. Additionally, we introduce a novel SHAP+LIME explainability method, accurately identifying the cancerous region in 97% of cases, ensuring model reliability for real-world use.

eess.IV↗

arXiv2023

A New Similarity Function for Spectral Clustering with Application to Plant Phenotypic Data

Clustering species of the same plant into different groups is an important step in developing new species of the concerned plant. Phenotypic (or physical) characteristics of plant species are commonly used to perform clustering. Hierarchical Clustering (HC) is popularly used for this task, and this algorithm suffers from low accuracy. In one of the recent works (Shastri et al., 2021), the authors have used the standard Spectral Clustering (SC) algorithm to improve the clustering accuracy. They have demonstrated the efficacy of their algorithm on soybean species. In the SC algorithm, one of the crucial steps is building the similarity matrix. A Gaussian similarity function is the standard choice to build this matrix. In the past, many works have proposed variants of the Gaussian similarity function to improve the performance of the SC algorithm, however, all have focused on the variance or scaling of the Gaussian. None of the past works have investigated upon the choice of base "e" (Euler's number) of the Gaussian similarity function (natural exponential function). Based upon spectral graph theory, specifically the Cheeger's inequality, in this work we propose use of a base "a" exponential function as the similarity function. We also integrate this new approach with the notion of "local scaling" from one of the first works that experimented with the scaling of the Gaussian similarity function (Zelnik-Manor et al., 2004). Using an eigenvalue analysis, we theoretically justify that our proposed algorithm should work better than the existing one. With evaluation on 2376 soybean species and 1865 rice species, we experimentally demonstrate that our new SC is 35% and 11% better than the standard SC, respectively.

cs.LG↗

arXiv2022

SABMIS: Sparse approximation based blind multi-image steganography scheme

We hide grayscale secret images into a grayscale cover image, which is considered to be a challenging steganography problem. Our goal is to develop a steganography scheme with enhanced embedding capacity while preserving the visual quality of the stego-image as well as the extracted secret image, and ensuring that the stego-image is resistant to steganographic attacks. The novel embedding rule of our scheme helps to hide secret image sparse coefficients into the oversampled cover image sparse coefficients in a staggered manner. The stego-image is constructed by using ADMM to solve the LASSO formulation of the underlying minimization problem. Finally, the secret images are extracted from the constructed stego-image using the reverse of our embedding rule. Using these components together, to achieve the above mentioned competing goals, forms our most novel contribution. We term our scheme SABMIS (Sparse Approximation Blind Multi-Image Steganography). We perform extensive experiments on several standard images. By choosing the size of the secret images to be half of the of cover image, we obtain embedding capacities of 2 bpp (bits per pixel), 4 bpp, 6 bpp, and 8 bpp while embedding one, two, three, and four secret images, respectively. Our focus is on hiding multiple secret images. For the case of hiding two and three secret images, our embedding capacities are higher than all the embedding capacities obtained in the literature until now. For the case of hiding four secret images, although our capacity is slightly lower than one work, we do better on the other two goals; a) very little deterioration in the quality of the stego-images and extracted secret images, and b) inherently and designed-to-be resistant to steganographic attacks. Additionally, we demonstrate that SABMIS executes in few minutes, and show its application on two real-life problems.

cs.CR↗

arXiv2021

Cube Sampled K-Prototype Clustering for Featured Data

Clustering large amount of data is becoming increasingly important in the current times. Due to the large sizes of data, clustering algorithm often take too much time. Sampling this data before clustering is commonly used to reduce this time. In this work, we propose a probabilistic sampling technique called cube sampling along with K-Prototype clustering. Cube sampling is used because of its accurate sample selection. K-Prototype is most frequently used clustering algorithm when the data is numerical as well as categorical (very common in today's time). The novelty of this work is in obtaining the crucial inclusion probabilities for cube sampling using Principal Component Analysis (PCA). Experiments on multiple datasets from the UCI repository demonstrate that cube sampled K-Prototype algorithm gives the best clustering accuracy among similarly sampled other popular clustering algorithms (K-Means, Hierarchical Clustering (HC), Spectral Clustering (SC)). When compared with unsampled K-Prototype, K-Means, HC and SC, it still has the best accuracy with the added advantage of reduced computational complexity (due to reduced data size).

cs.LG↗

arXiv2021

Resource Availability in the Social Cloud: An Economics Perspective

This paper focuses on social cloud formation, where agents are involved in a closeness-based conditional resource sharing and build their resource sharing network themselves. The objectives of this paper are: (1) to investigate the impact of agents' decisions of link addition and deletion on their local and global resource availability, (2) to analyze spillover effects in terms of the impact of link addition between a pair of agents on others' utility, (3) to study the role of agents' closeness in determining what type of spillover effects these agents experience in the network, and (4) to model the choices of agents that suggest with whom they want to add links in the social cloud. The findings include the following. Firstly, agents' decision of link addition (deletion) increases (decreases) their local resource availability. However, these observations do not hold in the case of global resource availability. Secondly, in a connected network, agents experience either positive or negative spillover effect and there is no case with no spillover effects. Agents observe no spillover effects if and only if the network is disconnected and consists of more than two components (sub-networks). Furthermore, if there is no change in the closeness of an agent (not involved in link addition) due to a newly added link, then the agent experiences negative spillover effect. Although an increase in the closeness of agents is necessary in order to experience positive spillover effects, the condition is not sufficient. By focusing on parameters such as closeness and shortest distances, we provide conditions under which agents choose to add links so as to maximise their resource availability.

cs.GT↗

arXiv2021

A Critical Note on Social Cloud

The idea of a social cloud has emerged as a resource sharing paradigm in a social network context. Undoubtedly, state-of-the-art social cloud systems demonstrate the potential of the social cloud acting as complementary to other computing paradigms such as the cloud, grid, peer-to-peer and volunteer computing. However, in this note, we have done a critical survey of the social cloud literature and come to the conclusion that these initial efforts fail to offer a general framework of the social cloud, also, to show the uniqueness of the social cloud. This short note reveals that there are significant differences regarding the concept of social cloud, resource definition, resource sharing and allocation mechanism, and its application and stakeholders. This study is an attempt to express a need for a general framework of the social cloud, which can incorporate various views and resource sharing setups discussed in the literature.

cs.DC↗

arXiv2021

CSIS: compressed sensing-based enhanced-embedding capacity image steganography scheme

Image steganography plays a vital role in securing secret data by embedding it in the cover images. Usually, these images are communicated in a compressed format. Existing techniques achieve this but have low embedding capacity. Enhancing this capacity causes a deterioration in the visual quality of the stego-image. Hence, our goal here is to enhance the embedding capacity while preserving the visual quality of the stego-image. We also intend to ensure that our scheme is resistant to steganalysis attacks. This paper proposes a Compressed Sensing Image Steganography (CSIS) scheme to achieve our goal while embedding binary data in images. The novelty of our scheme is the combination of three components in attaining the above-listed goals. First, we use compressed sensing to sparsify cover image block-wise, obtain its linear measurements, and then uniquely select permissible measurements. Further, before embedding the secret data, we encrypt it using the Data Encryption Standard (DES) algorithm, and finally, we embed two bits of encrypted data into each permissible measurement. Second, we propose a novel data extraction technique, which is lossless and completely recovers our secret data. Third, for the reconstruction of the stego-image, we use the least absolute shrinkage and selection operator (LASSO) for the resultant optimization problem. We perform experiments on several standard grayscale images and a color image, and evaluate embedding capacity, PSNR value, mean SSIM index, NCC coefficients, and entropy. We achieve 1.53 times more embedding capacity as compared to the most recent scheme. We obtain an average of 37.92 dB PSNR value, and average values close to 1 for both the mean SSIM index and the NCC coefficients, which are considered good. Moreover, the entropy of cover images and their corresponding stego-images are nearly the same.

cs.MM↗

arXiv2020

ParaLarH: Parallel FPGA Router based upon Lagrange Heuristics

Routing of the nets in Field Programmable Gate Array (FPGA) design flow is one of the most time consuming steps. Although Versatile Place and Route (VPR), which is a commonly used algorithm for this purpose, routes effectively, it is slow in execution. One way to accelerate this design flow is to use parallelization. Since VPR is intrinsically sequential, a set of parallel algorithms have been recently proposed for this purpose (ParaLaR and ParaLarPD). These algorithms formulate the routing process as a Linear Program (LP) and solve it using the Lagrange relaxation, the sub-gradient method, and the Steiner tree algorithm. Out of the many metrics available to check the effectiveness of routing, ParaLarPD, which is an improved version of ParaLaR, suffers from large violations in the constraints of the LP problem (which is related to the minimum channel width metric) as well as an easily measurable critical path delay metric that can be improved further. In this paper, we introduce a set of novel Lagrange heuristics that improve the Lagrange relaxation process. When tested on the MCNC benchmark circuits, on an average, this leads to halving of the constraints violation, up to 10% improvement in the minimum channel width, and up to 8% reduction in the critical path delay as obtained from ParaLarPD. We term our new algorithm as ParaLarH. Due to the increased work in the Lagrange relaxation process, as compared to ParaLarPD, ParaLarH does slightly deteriorate the speedup obtained because of parallelization, however, this aspect is easily compensated by using more number of threads.

cs.DC↗

arXiv2020

Probabilistically Sampled and Spectrally Clustered Plant Genotypes using Phenotypic Characteristics

Clustering genotypes based upon their phenotypic characteristics is used to obtain diverse sets of parents that are useful in their breeding programs. The Hierarchical Clustering (HC) algorithm is the current standard in clustering of phenotypic data. This algorithm suffers from low accuracy and high computational complexity issues. To address the accuracy challenge, we propose the use of Spectral Clustering (SC) algorithm. To make the algorithm computationally cheap, we propose using sampling, specifically, Pivotal Sampling that is probability based. Since application of samplings to phenotypic data has not been explored much, for effective comparison, another sampling technique called Vector Quantization (VQ) is adapted for this data as well. VQ has recently given promising results for genome data. The novelty of our SC with Pivotal Sampling algorithm is in constructing the crucial similarity matrix for the clustering algorithm and defining probabilities for the sampling technique. Although our algorithm can be applied to any plant genotypes, we test it on the phenotypic data obtained from about 2400 Soybean genotypes. SC with Pivotal Sampling achieves substantially more accuracy (in terms of Silhouette Values) than all the other proposed competitive clustering with sampling algorithms (i.e. SC with VQ, HC with Pivotal Sampling, and HC with VQ). The complexities of our SC with Pivotal Sampling algorithm and these three variants are almost same because of the involved sampling. In addition to this, SC with Pivotal Sampling outperforms the standard HC algorithm in both accuracy and computational complexity. We experimentally show that we are up to 45% more accurate than HC in terms of clustering accuracy. The computational complexity of our algorithm is more than a magnitude lesser than HC.

cs.LG↗

arXiv2020

Reusing Preconditioners in Projection based Model Order Reduction Algorithms

Dynamical systems are pervasive in almost all engineering and scientific applications. Simulating such systems is computationally very intensive. Hence, Model Order Reduction (MOR) is used to reduce them to a lower dimension. Most of the MOR algorithms require solving large sparse sequences of linear systems. Since using direct methods for solving such systems does not scale well in time with respect to the increase in the input dimension, efficient preconditioned iterative methods are commonly used. In one of our previous works, we have shown substantial improvements by reusing preconditioners for the parametric MOR (Singh et al. 2019). Here, we had proposed techniques for both, the non-parametric and the parametric cases, but had applied them only to the latter. We have four main contributions here. First, we demonstrate that preconditioners can be reused more effectively in the non-parametric case as compared to the parametric one because of the lack of parameters in the former. Second, we show that reusing preconditioners is an art and it needs to be fine-tuned for the underlying MOR algorithm. Third, we describe the pitfalls in the algorithmic implementation of reusing preconditioners. Fourth, and final, we demonstrate this theory on a real life industrial problem (of size 1.2 million), where savings of upto 64% in the total computation time is obtained by reusing preconditioners. In absolute terms, this leads to a saving of 5 days.

math.NA↗

arXiv2019

Externalities in Endogenous Sharing Economy Networks

This paper investigates the impact of link formation between a pair of agents on the resource availability of other agents (that is, externalities) in a social cloud network, a special case of endogenous sharing economy networks. Specifically, we study how the closeness between agents and the network size affect externalities. We conjecture, and experimentally support, that for an agent to experience positive externalities, an increase in its closeness is necessary. The condition is not sufficient though. We, then, show that for populated ring networks, one or more agents experience positive externalities due to an increase in the closeness of agents. Further, the initial distance between agents forming a link has a direct bearing on the number of beneficiaries, and the number of beneficiaries is always less than that of non-beneficiaries.

cs.GT↗

arXiv2019

Inexact Linear Solves In Model Reduction of Bilinear Dynamical Systems

Bilinear dynamical systems are commonly used in science and engineering because they form a bridge between linear and non-linear systems. However, simulating them is still a challenge because of their large size. Hence, a lot of research is currently being done for reducing such bilinear dynamical systems (termed as bilinear model order reduction or bilinear MOR). Bilinear iterative rational Krylov algorithm (BIRKA) is a very popular, standard and mathematically sound algorithm for bilinear MOR, which is based upon interpolatory projection technique. An efficient variant of BIRKA, Truncated BIRKA (or TBIRKA) has also been recently proposed. Like for any MOR algorithm, these two algorithms also require solving multiple linear systems as part of the model reduction process. For reducing very large dynamical systems, which is now-a-days becoming a norm, scaling of such linear systems with respect to input dynamical system size is a bottleneck. For efficiency, these linear systems are often solved by an iterative solver, which introduces approximation errors. Hence, stability analysis of MOR algorithms with respect to inexact linear solves is important. In our past work, we have shown that under mild conditions, BIRKA is stable (in the sense as discussed above). Here, we look at stability of TBIRKA in the same context. Besides deriving the conditions for a stable TBIRKA, our other novel contribution is the more intuitive methodology for achieving this. This approach exploits the fact that in TBIRKA a bilinear dynamical system can be represented by a finite set of functions, which was not possible in BIRKA (because infinite such functions were needed there). The stability analysis techniques that we propose here can be extended to many other methods for doing MOR of bilinear dynamical systems, e.g., using balanced truncation or the ADI methods.

math.NA↗

arXiv2018

Vector Quantized Spectral Clustering applied to Soybean Whole Genome Sequences

We develop a Vector Quantized Spectral Clustering (VQSC) algorithm that is a combination of Spectral Clustering (SC) and Vector Quantization (VQ) sampling for grouping Soybean genomes. The inspiration here is to use SC for its accuracy and VQ to make the algorithm computationally cheap (the complexity of SC is cubic in-terms of the input size). Although the combination of SC and VQ is not new, the novelty of our work is in developing the crucial similarity matrix in SC as well as use of k-medoids in VQ, both adapted for the Soybean genome data. We compare our approach with commonly used techniques like UPGMA (Un-weighted Pair Graph Method with Arithmetic Mean) and NJ (Neighbour Joining). Experimental results show that our approach outperforms both these techniques significantly in terms of cluster quality (up to 25% better cluster quality) and time complexity (order of magnitude faster).

q-bio.QM↗