Researcher profile

Sandeep Kumar

Sandeep Kumar contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author

28works

0followers

26topics

4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Are Data Augmentation and Segmentation Always Necessary? Insights from COVID-19 X-Rays and a Methodology Thereof

Purpose: Rapid and reliable diagnostic tools are crucial for managing respiratory diseases like COVID-19, where chest X-ray analysis coupled with artificial intelligence techniques has proven invaluable. However, most existing works on X-ray images have not considered lung segmentation, raising concerns about their reliability. Additionally, some have employed disproportionate and impractical augmentation techniques, making models less generalized and prone to overfitting. This study presents a critical analysis of both issues and proposes a methodology (SDL-COVID) for more reliable classification of chest X-rays for COVID-19 detection. Methods: We use class activation mapping to obtain a visual understanding of the predictions made by Convolutional Neural Networks (CNNs), validating the necessity of lung segmentation. To analyze the effect of data augmentation, deep learning models are implemented on two levels: one for an augmented dataset and another for a non-augmented dataset. Results: Careful analysis of X-ray images and their corresponding heat maps under expert medical supervision reveals that lung segmentation is necessary for accurate COVID-19 prediction. Regarding data augmentation, test accuracy significantly drops beyond a certain threshold with additional augmented images, indicating model overfitting. Conclusion: Our proposed methodology, SDL-COVID, achieves a precision of 95.21% and a lower false negative rate, ensuring its reliability for COVID-19 detection using chest X-rays.

preprint2026arXiv

ConstMig: Enabling Secure Live Migration of Large Intel SGX-based applications

Cloud service providers are adopting Trusted Execution Environments (TEEs) to provide hardware-guaranteed security to applications running on remote, untrusted data centers. However, migrating such applications still relies on the decade-old stop-and-copy method, which introduces large downtimes. Modern live-migration approaches such as pre-copy and post-copy do not work for TEE-based applications due to hardware-enforced restrictions. We propose ConstMig, a near-zero-downtime live-migration mechanism for large memory-footprint TEE-based applications. ConstMig is fully compatible with containers, virtual machines (VMs), and microVMs. Our prototype, built on Intel SGX, achieves near-zero downtime irrespective of enclave size and requires no additional hardware support. ConstMig reduces total downtime by 77 - 96% for a suite of SGX applications with multi-gigabyte memory footprints compared to state-of-the-art TEE-based migration solutions such as MigSGX.

preprint2026arXiv

GRAPHGINI: Fostering Individual and Group Fairness in Graph Neural Networks

Graph Neural Networks (GNNs) have demonstrated impressive performance across various tasks, leading to their increased adoption in high-stakes decision-making systems. However, concerns have arisen about GNNs potentially generating unfair decisions for underprivileged groups or individuals when lacking fairness constraints. This work addresses this issue by introducing GraphGini, a novel approach that incorporates the Gini coefficient to enhance both individual and group fairness within the GNN framework. We rigorously establish that the Gini coefficient offers greater robustness and promotes equal opportunity among GNN outcomes, advantages not afforded by the prevailing Lipschitz constant methodology. Additionally, we employ the Nash social welfare program to ensure our solution yields a Pareto optimal distribution of group fairness. Extensive experimentation on real-world datasets demonstrates GraphGini's efficacy in significantly improving individual fairness compared to state-of-the-art methods while maintaining utility and group fairness.

preprint2026arXiv

Quantum Gatekeeper: Multi-Factor Context-Bound Image Steganography with VQC Based Key Derivation on Quantum Hardware

This paper presents Quantum Gatekeeper, a context-bound image steganography framework where successful payload recovery depends on both cryptographic decryption and the reconstruction of a precise extraction path. The system integrates lossless least significant bit (LSB) embedding with a deterministic variational quantum circuit (VQC)-derived gate key, multi-factor contextual binding, and authenticated encryption. Payload extraction is contingent upon four requisite factors: a password, a shared secret, a user-supplied context string, and a reference image signature. Any deviation in these factors causes the system to read from an incorrect pixel sequence or fail authentication, resulting in silent rejection rather than partial disclosure. The proposed method derives a gatecontrolled extraction key from a seed-conditioned variational circuit, with parameters generated via cryptographic hash expansion and context-dependent image features. To ensure encode/decode consistency, the cryptographic key path is generated via exact statevector simulation; concurrently, IBM superconducting quantum hardware is utilized to evaluate the statistical behavior of the circuit family under physical noise. We introduce a dual-region image layout to resolve the nonce bootstrapping dependency, separating header recovery from payload recovery through independently derived keys. Experimental results confirm successful end-to-end message embedding and recovery on PNG images, demonstrating deterministic success under correct conditions and failure otherwise. The framework supports both text and image payloads; in the image-in-image configuration, a secret image is resized to a fixed resolution prior to embedding, enabling exact pixel-level recovery under correct contextual reconstruction.

preprint2026arXiv

QYOLO: Lightweight Object Detection via Quantum Inspired Shared Channel Mixing

The rapid advancement of object detection architectures has positioned single stage detectors as the dominant solution for real-time visual perception. A primary source of computational overhead in these models lies in the deep backbone stages, where C2f bottleneck modules at high stride levels accumulate a disproportionate share of parameters due to quadratic scaling with channel width. This work introduces QYOLO, a quantum-inspired channel mixing framework that achieves genuine architectural compression by replacing the two deepest backbone C2f modules at P4/16 (512 channels) and P5/32 (1024 channels) with a compact QMixBlock. The proposed block performs global channel recalibration through a sinusoidal mixing mechanism with shared learnable parameters across both backbone stages, enforcing consistent channel importance without requiring independent per-stage parameter sets. The neck and detection head remain fully classical and unchanged. Evaluation on the VisDrone2019 benchmark demonstrates that QYOLOv8n achieves a 20.2% reduction in parameter count (3.01M to 2.40M) and 12.3% GFLOPs reduction with only 0.4 pp mAP@50 degradation. QYOLOv8s achieves 21.8% reduction with 0.1 pp degradation. When combined with knowledge distillation, full accuracy parity is recovered at no cost to compression. An expanded backbone plus neck variant achieved 38 to 41% reduction at the cost of greater accuracy degradation, motivating the backbone-only final design.

preprint2026arXiv

When Reviews Disagree: Fine-Grained Contradiction Analysis in Scientific Peer Reviews

Scientific peer reviews frequently contain conflicting expert judgments, and the increasing scale of conference submissions makes it challenging for Area Chairs and editors to reliably identify and interpret such disagreements. Existing approaches typically frame reviewer disagreement as binary contradiction detection over isolated sentence pairs, abstracting away the review-level context and obscuring differences in the severity of evaluative conflict. In this work, we introduce a fine-grained formulation of reviewer contradiction analysis that operates over full peer reviews by explicitly identifying contradiction evidence spans and assigning graded disagreement intensity scores. To support this task, we present RevCI, an expert-annotated benchmark of peer-review pairs with evidence-level contradiction annotations with graded intensity labels. We further propose IMPACT, a structured multi-agent framework that integrates aspect-conditioned evidence extraction, deliberative reasoning, and adjudication to model reviewer contradictions and their intensity. To support efficient deployment, we distill IMPACT into TIDE, a small language model that predicts contradiction evidence and intensity in a single forward pass. Experimental results show that IMPACT substantially outperforms strong single-agent and generic multi-agent baselines in both evidence identification and intensity agreement, while TIDE achieves competitive performance at significantly lower inference cost.

preprint2022arXiv

A Comprehensive Benchmark Suite for Intel SGX

Trusted execution environments (TEEs) such as \intelsgx facilitate the secure execution of an application on untrusted machines. Sadly, such environments suffer from serious limitations and performance overheads in terms of writing back data to the main memory, their interaction with the OS, and the ability to issue I/O instructions. There is thus a plethora of work that focuses on improving the performance of such environments -- this necessitates the need for a standard, widely accepted benchmark suite (something similar to SPEC and PARSEC). To the best of our knowledge, such a suite does not exist. Our suite, SGXGauge, contains a diverse set of workloads such as blockchain codes, secure machine learning algorithms, lightweight web servers, secure key-value stores, etc. We thoroughly characterizes the behavior of the benchmark suite on a native platform and on a platform that uses a library OS-based shimming layer (GrapheneSGX). We observe that the most important metrics of interest are performance counters related to paging, memory, and TLB accesses. There is an abrupt change in performance when the memory footprint starts to exceed the size of the EPC size in Intel SGX, and the library OS does not add a significant overhead (~ +- 10%).

preprint2022arXiv

A New Monte-Carlo Radiative Transfer Simulation of Cyclotron Resonant Scattering Features

We present a new Monte-Carlo radiative transfer code, which we have used to model the cyclotron line features in the environment of a variable magnetic field and plasma density. The code accepts an input continuum and performs only the line transfer by including the three cyclotron resonant processes (cyclotron absorption, cyclotron emission, cyclotron scattering). Subsequently, the effects of gravitational red-shift and light bending on the emergent spectra are computed. We have applied our code to predict the observable spectra from three different emission geometries; 1) an optically thin slab near the stellar surface, 2) an accretion mound formed by the accumulation of the accreted matter, 3) an accretion column representing the zone of a settling flow onto the star. Our results show that the locally emergent spectra from the emission volume are significantly anisotropic. However, in the presence of strong light bending the anisotropy reduces considerably. This averaging also drastically reduces the strength of harmonics higher than second in the observable cyclotron spectra. We find that uniform field slabs produce line features that are too narrow, and mounds with large magnetic distortions produce features that are too wide compared to the average widths of the spectral features observed from various sources. The column with a gently varying (dipole) field produces widths in the intermediate range, similar to those observed.

preprint2022arXiv

Enhancement to Training of Bidirectional GAN : An Approach to Demystify Tax Fraud

Outlier detection is a challenging activity. Several machine learning techniques are proposed in the literature for outlier detection. In this article, we propose a new training approach for bidirectional GAN (BiGAN) to detect outliers. To validate the proposed approach, we train a BiGAN with the proposed training approach to detect taxpayers, who are manipulating their tax returns. For each taxpayer, we derive six correlation parameters and three ratio parameters from tax returns submitted by him/her. We train a BiGAN with the proposed training approach on this nine-dimensional derived ground-truth data set. Next, we generate the latent representation of this data set using the $encoder$ (encode this data set using the $encoder$) and regenerate this data set using the $generator$ (decode back using the $generator$) by giving this latent representation as the input. For each taxpayer, compute the cosine similarity between his/her ground-truth data and regenerated data. Taxpayers with lower cosine similarity measures are potential return manipulators. We applied our method to analyze the iron and steel taxpayers data set provided by the Commercial Taxes Department, Government of Telangana, India.

preprint2022arXiv

ErfAct and Pserf: Non-monotonic Smooth Trainable Activation Functions

An activation function is a crucial component of a neural network that introduces non-linearity in the network. The state-of-the-art performance of a neural network depends also on the perfect choice of an activation function. We propose two novel non-monotonic smooth trainable activation functions, called ErfAct and Pserf. Experiments suggest that the proposed functions improve the network performance significantly compared to the widely used activations like ReLU, Swish, and Mish. Replacing ReLU by ErfAct and Pserf, we have 5.68% and 5.42% improvement for top-1 accuracy on Shufflenet V2 (2.0x) network in CIFAR100 dataset, 2.11% and 1.96% improvement for top-1 accuracy on Shufflenet V2 (2.0x) network in CIFAR10 dataset, 1.0%, and 1.0% improvement on mean average precision (mAP) on SSD300 model in Pascal VOC dataset.

preprint2022arXiv

Flexoelectronic doping of the degenerate silicon and the correlated electron behavior

In metal/degenerately doped silicon bilayer structure, the interfacial flexoelectric effect due to strain gradient leads to charge carrier transfer from metal layer to the silicon layer. This excess charge carrier concentration is called flexoelectronic doping or flexoelectronic charge transfer, which gives rise to an electronically polarized (order of magnitude larger than ferroelectric materials) silicon layer. In the transport measurements, the charge carrier concentration in silicon is found to increase by two orders of magnitude due to flexoelectronic doping, which changes the Fermi level and the Hall response. The flexoelectronic charge accumulation modifies the electron-electron and the electron phonon coupling, which gives rise to Mott metal-insulator transition and magnetism of phonons, respectively. The coexistence of flexoelectronic polarization and magnetism gives rise to a new class of materials called electronic multiferroics. By controlling the flexoelectronic doping, material behavior can potentially be engineered for quantum, spintronics and electronics applications in semiconductor materials.

preprint2022arXiv

On the Integral and Derivative Identities of Bivariate Fox H-Function: Application in Wireless System Performance Analysis

The present work proposes analytical solutions for the integral of bivariate Fox H-function in combination with algebraic, exponential, and complementary error functions. In addition, the work also presents the derivative identities with respect to function arguments. Further, the suitability of the proposed mathematical solutions is verified with reference to wireless communication environment, where a fading behaviour of the channel acquired the bivariate Fox H-function structure. Further more, asymptotic results for the outage probability and average symbol error probability are presented utilizing the origin probability density function based approach. The obtained results are free from complex analytical functions. At last, the analytical findings of the paper are compared with the numerical results and also with the Monte-Carlo simulation results to confirm their accuracy.

preprint2022arXiv

Parallel Multiphysics Simulation for the Stabilized Optimal Transportation Meshfree (OTM)

This paper presents a parallel \PG{implementation} for the Optimal Transportation Meshfree (OTM) method on large CPU clusters. Communications are handled with the Message Passing Interface (MPI). The Recursive Coordinate Bisection (RCB) algorithm is utilized for domain decomposition and for implementing dynamic load-balancing strategy. This work involves three new concepts to reduce the computational efforts: Dynamic halo regions, Efficient data management strategies for ease of addition and deletion of nodes and material points using advanced STL container, and nearest neighborhood communication for detection of neighbors and communication. Also, Linked Cell approach has been implemented to further reduce the computational efforts. Parallel performance analysis is investigated for challenging multiphysics applications like Taylor rod impact and serrated chip formation process. Adequate scalability of parallel implementation for these applications is reported.

preprint2022arXiv

Probing of large interfacial contribution to spin orbit coupling in CoFeB/Ta heterostructure by ultrafast THz emission spectroscopy

Ultrafast THz radiation generation from ferromagnetic/nonmagnetic bilayer heterostructure-based spintronic emitters generally exploits the conversion from spin- to charge-current within the nonmagnetic layer and its interface with the ferromagnetic layer. Various possible sub-contributions to the underlying mechanism of inverse spin Hall effect for the THz emission from such structures, need to be exploited for not only investigating the intricacies at the fundamental level in the material properties themselves but also for improving their performance for broadband and high-power THz emission. Here, we report ultrafast THz emission from CoFeB/Ta bilayer at varying sample temperatures in a large range to unravel the role of intrinsic and extrinsic spin to charge conversion processes. In addition to an enhancement in the THz emission, its temperature dependence shows a THz signal polarity reversal if the CoFeB/Ta sample is annealed at an elevated temperature. We extract the behaviour of the spin Hall resistivity, determine the intrinsic spin Hall conductivity contribution in it and compare those with the standard Fe/Pt system. Our results clearly demonstrate a giant interfacial contribution to the overall spin Hall angle arising from the modified interface in the annealed CoFeB/Ta, where a sign reversal in the corresponding spin Hall angle is manifested from the THz amplitude variation with the temperature.

preprint2022arXiv

Robust Graph Neural Networks using Weighted Graph Laplacian

Graph neural network (GNN) is achieving remarkable performances in a variety of application domains. However, GNN is vulnerable to noise and adversarial attacks in input data. Making GNN robust against noises and adversarial attacks is an important problem. The existing defense methods for GNNs are computationally demanding and are not scalable. In this paper, we propose a generic framework for robustifying GNN known as Weighted Laplacian GNN (RWL-GNN). The method combines Weighted Graph Laplacian learning with the GNN implementation. The proposed method benefits from the positive semi-definiteness property of Laplacian matrix, feature smoothness, and latent features via formulating a unified optimization framework, which ensures the adversarial/noisy edges are discarded and connections in the graph are appropriately weighted. For demonstration, the experiments are conducted with Graph convolutional neural network(GCNN) architecture, however, the proposed framework is easily amenable to any existing GNN architecture. The simulation results with benchmark dataset establish the efficacy of the proposed method, both in accuracy and computational efficiency. Code can be accessed at https://github.com/Bharat-Runwal/RWL-GNN.

preprint2022arXiv

SMU: smooth activation function for deep networks using smoothing maximum technique

Deep learning researchers have a keen interest in proposing two new novel activation functions which can boost network performance. A good choice of activation function can have significant consequences in improving network performance. A handcrafted activation is the most common choice in neural network models. ReLU is the most common choice in the deep learning community due to its simplicity though ReLU has some serious drawbacks. In this paper, we have proposed a new novel activation function based on approximation of known activation functions like Leaky ReLU, and we call this function Smooth Maximum Unit (SMU). Replacing ReLU by SMU, we have got 6.22% improvement in the CIFAR100 dataset with the ShuffleNet V2 model.

preprint2022arXiv

Static and Dynamical, Fractional Uncertainty Principles

We study the process of dispersion of low-regularity solutions to the Schrödinger equation using fractional weights (observables). We give another proof of the uncertainty principle for fractional weights and use it to get a lower bound for the concentration of mass. We consider also the evolution when the initial datum is the Dirac comb in $\mathbb{R}$. In this case we find fluctuations that concentrate at rational times and that resemble a realization of a Lévy process. Furthermore, the evolution exhibits multifractality.

preprint2021arXiv

Closed-loop targeted optogenetic stimulation of C. elegans populations

We present a high-throughput optogenetic illumination system capable of simultaneous closed-loop light delivery to specified targets in populations of moving Caenorhabditis elegans. The instrument addresses three technical challenges: it delivers targeted illumination to specified regions of the animal's body such as its head or tail; it automatically delivers stimuli triggered upon the animal's behavior; and it achieves high throughput by targeting many animals simultaneously. The instrument was used to optogenetically probe the animal's behavioral response to competing mechanosensory stimuli in the the anterior and posterior soft touch receptor neurons. Responses to more than $10^4$ stimulus events from a range of anterior-posterior intensity combinations were measured. The animal's probability of sprinting forward in response to a mechanosensory stimulus depended on both the anterior and posterior stimulation intensity, while the probability of reversing depended primarily on the posterior stimulation intensity. We also probed the animal's response to mechanosensory stimulation during the onset of turning, a relatively rare behavioral event, by delivering stimuli automatically when the animal began to turn. Using this closed-loop approach, over $10^3$ stimulus events were delivered during turning onset at a rate of 9.2 events per worm-hour, a greater than 25-fold increase in throughput compared to previous investigations. These measurements validate with greater statistical power previous findings that turning acts to gate mechanosensory evoked reversals. Compared to previous approaches, the current system offers targeted optogenetic stimulation to specific body regions or behaviors with many-fold increases in throughput to better constrain quantitative models of sensorimotor processing.

preprint2020arXiv

A Novel GDP Prediction Technique based on Transfer Learning using CO2 Emission Dataset

In the last 150 years, CO2 concentration in the atmosphere has increased from 280 parts per million to 400 parts per million. This has caused an increase in the average global temperatures by nearly 0.7 degree centigrade due to the greenhouse effect. However, the most prosperous states are the highest emitters of greenhouse gases (specially, CO2). This indicates a strong relationship between gaseous emissions and the gross domestic product (GDP) of the states. Such a relationship is highly volatile and nonlinear due to its dependence on the technological advancements and constantly changing domestic and international regulatory policies and relations. To analyse such vastly nonlinear relationships, soft computing techniques has been quite effective as they can predict a compact solution for multi-variable parameters without any explicit insight into the internal system functionalities. This paper reports a novel transfer learning based approach for GDP prediction, which we have termed as Domain Adapted Transfer Learning for GDP Prediction. In the proposed approach per capita GDP of different nations is predicted using their CO2 emissions via a model trained on the data of any developed or developing economy. Results are comparatively presented considering three well-known regression methods such as Generalized Regression Neural Network, Extreme Learning Machine and Support Vector Regression. Then the proposed approach is used to reliably estimate the missing per capita GDP of some of the war-torn and isolated countries.

preprint2020arXiv

Deep Bayesian Network for Visual Question Generation

Generating natural questions from an image is a semantic task that requires using vision and language modalities to learn multimodal representations. Images can have multiple visual and language cues such as places, captions, and tags. In this paper, we propose a principled deep Bayesian learning framework that combines these cues to produce natural questions. We observe that with the addition of more cues and by minimizing uncertainty in the among cues, the Bayesian network becomes more confident. We propose a Minimizing Uncertainty of Mixture of Cues (MUMC), that minimizes uncertainty present in a mixture of cues experts for generating probabilistic questions. This is a Bayesian framework and the results show a remarkable similarity to natural questions as validated by a human study. We observe that with the addition of more cues and by minimizing uncertainty among the cues, the Bayesian framework becomes more confident. Ablation studies of our model indicate that a subset of cues is inferior at this task and hence the principled fusion of cues is preferred. Further, we observe that the proposed approach substantially improves over state-of-the-art benchmarks on the quantitative metrics (BLEU-n, METEOR, ROUGE, and CIDEr). Here we provide project link for Deep Bayesian VQG \url{https://delta-lab-iitk.github.io/BVQG/}

preprint2020arXiv

Flexoelectric effect mediated spin-to-charge conversion at amorphous-Si thin film interfaces

Interfacial spin to charge conversion arises due to an electric potential perpendicular to the interface. The electric potential can be artificially induced, for example, using ferroelectric and piezoelectric thin films at the interface. An alternate way to induce the electric potential could be flexoelectric field. The flexoelectricity can be observed in all the material that either have or lack inversion symmetry, additionally no large gate bias is needed. In this experimental study, we report large spin to charge conversion (spin-Hall angle- 0.578) at Ni80Fe20/amorphous-Si interfaces attributed to flexoelectricity mediated Rashba spin-orbit coupling. The flexoelectricity at the interface also gave rise to interlayer spin-acoustic phonon or flexo-magnetoelastic coupling. In addition to spin-charge conversion, the strained interfaces also led to almost three-fold increase in anomalous Nernst effect. This strain engineering for spin dependent thermoelectric behavior at room temperature opens a new window to the realization of spintronics and spin-caloritronics devices.

preprint2020arXiv

Group ring based public key cryptosystems

In this paper, we propose two cryptosystems based on group rings and existing cryptosystem. First one is Elliptic ElGamal type group ring public key cryptosystem whose security is greater than security of cryptosystems based on elliptic curves discrete logarithmic problem (ECDLP). Second is ElGamal type group ring public key cryptosystem, which is analogous to ElGamal public key cryptosystem but has comparatively greater security. Examples are also given for both the proposed cryptosystems.

preprint2020arXiv

Interlacing properties of system-poles, system-zeros and spectral-zeros in MIMO systems

SISO passive systems with just one type of memory/storage element (either only inductive or only capacitative) are known to have real poles and zeros, and further, with the zeros interlacing poles (ZIP). Due to a variety of definitions of the notion of a system zero, and due to other reasons described in the paper, results involving ZIP have not been extended to MIMO systems. This paper formulates conditions under which MIMO systems too have interlaced poles and zeros. This paper next focusses on the notion of a `spectral zero' of a system, which has been well-studied in various contexts: for example, spectral factorization, optimal charging/discharging of a dissipative system, and even model order reduction. We formulate conditions under which the spectral zeros of a MIMO system are real, and further, conditions that guarantee that the system-zeros, spectral zeros, and the poles are all interlaced. The techniques used in the proofs involve new results in Algebraic Riccati equations (ARE) and Hamiltonian matrices, and these results help in formulating new notions of positive-real balancing, and inter-relations with the existing notion of positive-real balancing; we also relate the positive-real singular values with the eigenvalues of the extremal ARE solutions in the proposed `quasi-balanced' forms.

preprint2020arXiv

Large spin-Hall effect in Si at room temperature

Silicon's weak intrinsic spin-orbit coupling and centrosymmetric crystal structure are a critical bottleneck to the development of Si spintronics, because they lead to an insignificant spin-Hall effect (spin current generation) and inverse spin-Hall effect (spin current detection). Here, we undertake current, magnetic field, crystallography dependent magnetoresistance and magneto thermal transport measurements to study the spin transport behavior in freestanding Si thin films. We observe a large spin-Hall magnetoresistance in both p-Si and n-Si at room temperature and it is an order of magnitude larger than that of Pt. One explanation of the unexpectedly large and efficient spin-Hall effect is spin-phonon coupling instead of spin-orbit coupling. The macroscopic origin of the spin-phonon coupling can be large strain gradients that can exist in the freestanding Si films. This discovery in a light, earth abundant and centrosymmetric material opens a new path of strain engineering to achieve spin dependent properties in technologically highly-developed materials.

preprint2020arXiv

Magnetic field-dependent resistance crossover and logarithmic to non-saturating magnetoresistance in topological insulator Bi$_2$Te$_3$

We report a metal-insulator like transition in single crystalline 3D topological insulator Bi2Te3 at a temperature of 230K in presence of an external magnetic field applied normal to the surface. This transition becomes more prominent at larger magnetic field strength with the residual resistance value increasing linearly with the magnetic field. At low temperature, the magnetic field dependence of the magnetoresistance shows a transition from logarithmic to linear behavior and the onset magnetic field value for this transition decreases with increasing temperature. The logarithmic magnetoresistance indicates the weak anti-localization of the surface Dirac electrons while the high temperature behavior originates from the bulk carriers due to intrinsic impurities. At even higher temperatures beyond~230 K, a completely classical Lorentz model type quadratic behavior of the magnetoresistance is observed. We also show that the experimentally observed anomalies at ~230K in the magneto-transport properties do not originate from any stacking fault in Bi2Te3.

preprint2020arXiv

Photo-Seebeck effect in single-crystalline bismuth telluride topological insulator

Bismuth telluride is a low energy bulk band-gap topological system with conducting surface states. Besides its very good thermoelectric properties, it also makes a very good candidate for broadband photodetectors. Here, we report temperature-dependent photo-Seebeck effect in a bulk single crystalline bismuth telluride. On light illumination, an electrically biased sample shows distinguishable contributions in the measured current due to both the Seebeck effect and the normal photo-generated carriers within a narrow layer of the sample. Detailed experiments are performed to elucidate the distinction between the Seebeck contribution and the photogenerated current. The temperature-dependence of the photocurrent without Seebeck contribution shows a sign reversal from negative to positive at a specific temperature depending on the wavelength of photoexcitation light.

preprint2020arXiv

TanhSoft -- a family of activation functions combining Tanh and Softplus

Deep learning at its core, contains functions that are composition of a linear transformation with a non-linear function known as activation function. In past few years, there is an increasing interest in construction of novel activation functions resulting in better learning. In this work, we propose a family of novel activation functions, namely TanhSoft, with four undetermined hyper-parameters of the form tanh(αx+βe^{γx})ln(δ+e^x) and tune these hyper-parameters to obtain activation functions which are shown to outperform several well known activation functions. For instance, replacing ReLU with xtanh(0.6e^x)improves top-1 classification accuracy on CIFAR-10 by 0.46% for DenseNet-169 and 0.7% for Inception-v3 while with tanh(0.87x)ln(1 +e^x) top-1 classification accuracy on CIFAR-100 improves by 1.24% for DenseNet-169 and 2.57% for SimpleNet model.

preprint2019arXiv

Meandering gate edges for breakdown voltage enhancement in AlGaN/GaN HEMTs

In this letter, we report on a unique device design strategy for increasing the breakdown voltage and hence Baliga Figure of Merit (BFOM) of III-nitride HEMTs by engineering the gate edge towards the drain. The breakdown of such devices with meandering gate-drain access region (M-HEMT) are found to be 62% more compared to that of conventional HEMT while the ON resistance suffers by 76%, leading to an overall improvement in the BFOM for by 28%. 3D-TCAD simulations show that the decrease in the peak electric field at the gate edge was responsible for increased breakdown voltage.

Sandeep Kumar

Quick read

Decide how to stay connected

How to connect with this researcher

Open a focused conversation when the fit is right

See the researcher in context

Building this graph slice

28 published item(s)

Are Data Augmentation and Segmentation Always Necessary? Insights from COVID-19 X-Rays and a Methodology Thereof

ConstMig: Enabling Secure Live Migration of Large Intel SGX-based applications

GRAPHGINI: Fostering Individual and Group Fairness in Graph Neural Networks

Quantum Gatekeeper: Multi-Factor Context-Bound Image Steganography with VQC Based Key Derivation on Quantum Hardware

QYOLO: Lightweight Object Detection via Quantum Inspired Shared Channel Mixing

When Reviews Disagree: Fine-Grained Contradiction Analysis in Scientific Peer Reviews

A Comprehensive Benchmark Suite for Intel SGX

A New Monte-Carlo Radiative Transfer Simulation of Cyclotron Resonant Scattering Features

Enhancement to Training of Bidirectional GAN : An Approach to Demystify Tax Fraud

ErfAct and Pserf: Non-monotonic Smooth Trainable Activation Functions

Flexoelectronic doping of the degenerate silicon and the correlated electron behavior

On the Integral and Derivative Identities of Bivariate Fox H-Function: Application in Wireless System Performance Analysis

Parallel Multiphysics Simulation for the Stabilized Optimal Transportation Meshfree (OTM)

Probing of large interfacial contribution to spin orbit coupling in CoFeB/Ta heterostructure by ultrafast THz emission spectroscopy

Robust Graph Neural Networks using Weighted Graph Laplacian

SMU: smooth activation function for deep networks using smoothing maximum technique

Static and Dynamical, Fractional Uncertainty Principles

Closed-loop targeted optogenetic stimulation of C. elegans populations

A Novel GDP Prediction Technique based on Transfer Learning using CO2 Emission Dataset

Deep Bayesian Network for Visual Question Generation

Flexoelectric effect mediated spin-to-charge conversion at amorphous-Si thin film interfaces

Group ring based public key cryptosystems

Interlacing properties of system-poles, system-zeros and spectral-zeros in MIMO systems

Large spin-Hall effect in Si at room temperature

Magnetic field-dependent resistance crossover and logarithmic to non-saturating magnetoresistance in topological insulator Bi$_2$Te$_3$

Photo-Seebeck effect in single-crystalline bismuth telluride topological insulator

TanhSoft -- a family of activation functions combining Tanh and Softplus

Meandering gate edges for breakdown voltage enhancement in AlGaN/GaN HEMTs