Searcharxiv⌕ Search

arXiv subjects

Ayan Banerjee

Publications and source records attributed to Ayan Banerjee.

At least 91 records · Page 5Linked to original sources
arXiv2024

Investigating Stable Quark Stars in Rastall-Rainbow Gravity and Their Compatibility with Gravitational Wave Observations

We present a stable model for quark stars in Rastall-Rainbow (R-R) gravity. The structure of this configuration is obtained by utilizing an interacting quark matter equation of state. The R-R gravity theory is developed as a combination of two distinct theories, namely, the Rastall theory and the gravity's rainbow formalism. Depending on the model parameters ($\barλ, η, Σ$), the mass-radius relations are numerically computed for modified Tolman-Oppenheimer-Volkoff (TOV) equations with proper boundary conditions. The stability of equilibrium configuration has been checked through the static stability criterion, adiabatic index and the sound velocity. Our calculations predict larger maximum masses for quark stars, and the obtained results are compatible with accepted masses and radii values, including constraints from GW190814 and GW170817 events in all the studied cases.

gr-qc↗
arXiv2024

GraphKD: Exploring Knowledge Distillation Towards Document Object Detection with Structured Graph Creation

Object detection in documents is a key step to automate the structural elements identification process in a digital or scanned document through understanding the hierarchical structure and relationships between different elements. Large and complex models, while achieving high accuracy, can be computationally expensive and memory-intensive, making them impractical for deployment on resource constrained devices. Knowledge distillation allows us to create small and more efficient models that retain much of the performance of their larger counterparts. Here we present a graph-based knowledge distillation framework to correctly identify and localize the document objects in a document image. Here, we design a structured graph with nodes containing proposal-level features and edges representing the relationship between the different proposal regions. Also, to reduce text bias an adaptive node sampling strategy is designed to prune the weight distribution and put more weightage on non-text nodes. We encode the complete graph as a knowledge representation and transfer it from the teacher to the student through the proposed distillation loss by effectively capturing both local and global information concurrently. Extensive experimentation on competitive benchmarks demonstrates that the proposed framework outperforms the current state-of-the-art approaches. The code will be available at: https://github.com/ayanban011/GraphKD.

cs.CV↗
arXiv2024

Inhomogeneous spin momentum induced orbital motion of birefringent particles in tight focusing of vector beams in optical tweezers

Spin orbit interaction (SOI) due to tight focusing of light in optical tweezers has led to exciting and exotic avenues towards inducing rotation in microscopic particles. However, instances where the back action of the particles influences and modifies SOI effects so as to induce rotational motion are rarely known. Here, we tightly focus a vector beam having radial/azimuthal polarization carrying no intrinsic angular momentum, into a refractive index stratified medium, and observe orbital rotation of birefringent particles around the beam propagation axis. In order to validate our experimental findings, we perform numerical simulations of the underlying equations. Our simulations reveal that the interaction of light with a birefringent particle gives rise to inhomogeneous spin currents near the focus, resulting in a finite spin momentum. This spin momentum combines with the canonical momentum to finally generate an origin-dependent orbital angular momentum which is manifested in the rotation of the birefringent particles around the beam axis. Our study describes a unique modulation of the SOI of light due to interaction with anisotropic particles that can be used to identify new avenues for exotic and complex particle manipulation in optical tweezers.

physics.optics↗
arXiv2024

Observation of Larmor-like precession of a single birefringent particle due to spin-dependent forces in tilted optical tweezers

We observe clear precessional motion of highly birefringent liquid crystal (LC) particles trapped in a spherically aberrated optical trap which is built around a tilted refractive index stratified medium. For input circularly polarized light, the breaking of azimuthal symmetry induced by the tilt leads to an asymmetric intensity distribution in the radial direction near the trap focal plane, which - in combination with the spin-orbit conversion effects for input circularly polarized light - results in non-uniform canonical and spin momentum densities in those regions. In addition, while the canonical momentum remains always oriented towards the axial direction, the spin momentum reverses direction along spatial loops in the radial direction. As a consequence, the total momentum precesses around the canonical momentum vector along elliptical spatial loops - akin to a Larmor-like precession of magnetic moment (total momentum in our case) around a magnetic field (canonical momentum). We probe this precession experimentally using the single trapped LC particles - with the direction of precession determined by the helicity of the input light and the precession frequency varying linearly with the laser power. Our experimental results are validated by numerical simulations of the system where we employ the Debye-Wolf theory for tight focusing in the presence of a tilted stratified media.

physics.optics↗
arXiv2023

Effects of anisotropic pressure on interacting quark star structure

Perturbative Quantum Chromodynamics (pQCD) corrections and color superconductivity predict that strongly interacting matter can reveal new physical phenomena under extreme conditions. Taking into account these interaction effects, we investigate the role of anisotropic pressure in quark stars composed of interacting quark matter. Adopting two physically well-motivated anisotropy profiles, we numerically solve the stellar structure equations in order to explore the consequences of anisotropic pressure on various macroscopic properties such as radius, gravitational mass, surface redshift, moment of inertia, tidal Love number and oscillation spectrum. Remarkably, for both anisotropy models, negative anisotropies increase the radial stability of interacting quark stars, while the opposite occurs for positive anisotropies. However, for the Bowers-Liang profile, the central density corresponding to the maximum-mass point does not coincide with the central density where the squared oscillation frequency vanishes, indicating that the existence of stable anisotropic interacting quark stars is possible beyond the maximum mass for negative anisotropies. Additionally, we compare our theoretical predictions with several observational mass-radius measurements and tidal deformability constraints, which suggest that both strong interaction effects and anisotropy effects play a crucial role in describing compact stars observed in the Universe.

gr-qc↗
arXiv2023

The Expert Knowledge combined with AI outperforms AI Alone in Seizure Onset Zone Localization using resting state fMRI

We evaluated whether integration of expert guidance on seizure onset zone (SOZ) identification from resting state functional MRI (rs-fMRI) connectomics combined with deep learning (DL) techniques enhances the SOZ delineation in patients with refractory epilepsy (RE), compared to utilizing DL alone. Rs-fMRI were collected from 52 children with RE who had subsequently undergone ic-EEG and then, if indicated, surgery for seizure control (n = 25). The resting state functional connectomics data were previously independently classified by two expert epileptologists, as indicative of measurement noise, typical resting state network connectivity, or SOZ. An expert knowledge integrated deep network was trained on functional connectomics data to identify SOZ. Expert knowledge integrated with DL showed a SOZ localization accuracy of 84.8& and F1 score, harmonic mean of positive predictive value and sensitivity, of 91.7%. Conversely, a DL only model yielded an accuracy of less than 50% (F1 score 63%). Activations that initiate in gray matter, extend through white matter and end in vascular regions are seen as the most discriminative expert identified SOZ characteristics. Integration of expert knowledge of functional connectomics can not only enhance the performance of DL in localizing SOZ in RE, but also lead toward potentially useful explanations of prevalent co-activation patterns in SOZ. RE with surgical outcomes and pre-operative rs-fMRI studies can yield expert knowledge most salient for SOZ identification.

cs.CV↗
arXiv2023

Detection of Unknown-Unknowns in Human-in-Plant Human-in-Loop Systems Using Physics Guided Process Models

Unknown-unknowns are operational scenarios in systems that are not accounted for in the design and test phase. In such scenarios, the operational behavior of the Human-in-loop (HIL) Human-in-Plant (HIP) systems is not guaranteed to meet requirements such as safety and efficacy. We propose a novel framework for analyzing the operational output characteristics of safety-critical HIL-HIP systems that can discover unknown-unknown scenarios and evaluate potential safety hazards. We propose dynamics-induced hybrid recurrent neural networks (DiH-RNN) to mine a physics-guided surrogate model (PGSM) that checks for deviation of the cyber-physical system (CPS) from safety-certified operational characteristics. The PGSM enables early detection of unknown-unknowns based on the physical laws governing the system. We demonstrate the detection of operational changes in an Artificial Pancreas(AP) due to unknown insulin cartridge errors.

cs.AI↗
arXiv2023

Harnessing the Power of Multi-Lingual Datasets for Pre-training: Towards Enhancing Text Spotting Performance

The adaptation capability to a wide range of domains is crucial for scene text spotting models when deployed to real-world conditions. However, existing state-of-the-art (SOTA) approaches usually incorporate scene text detection and recognition simply by pretraining on natural scene text datasets, which do not directly exploit the intermediate feature representations between multiple domains. Here, we investigate the problem of domain-adaptive scene text spotting, i.e., training a model on multi-domain source data such that it can directly adapt to target domains rather than being specialized for a specific domain or scenario. Further, we investigate a transformer baseline called Swin-TESTR to focus on solving scene-text spotting for both regular and arbitrary-shaped scene text along with an exhaustive evaluation. The results clearly demonstrate the potential of intermediate representations to achieve significant performance on text spotting benchmarks across multiple domains (e.g. language, synth-to-real, and documents). both in terms of accuracy and efficiency.

cs.CV↗
arXiv2023

Non-Commutative Effects on Wormholes in Rastall-Rainbow Gravity

In this present article, we explore the physical properties and characteristics of static, spherically symmetric wormholes in the background of Rastall-Rainbow gravity. The Rastall-Rainbow gravity theory has recently been proposed as a combination of two theories, namely, the Rastall theory and the Rainbow description. We implemented noncommutativity by adopting two different distributions of energy density (Gaussian and Lorentzian) in the Morris and Thorne metric. We solve the field equations analytically and discuss all the properties of wormholes depending on the two model parameters. Notably, for specific parameter ranges, one can alleviate the violation of the WEC at the throat and its neighbourhood.

gr-qc↗
arXiv2023

Non-Hermitian Topology and Flat Bands via an Exact Real Space Decimation Scheme

In recent years, non-Hermitian phases in classical and quantum systems have garnered significant attention. In particular, their intriguing band geometry offers a platform for exploring unique topological states and unconventional quantum dynamics. However, their topological characterization becomes particularly interesting and challenging in complex multiband systems. Here we propose a decimation framework, which leverages real space renormalization group to streamline the analysis of complex multiband non-Hermitian systems. Our systematic approach allows us to probe different phases and transitions, analyze bulk-boundary correspondence, formulate generalized Brillouin zones, investigate open boundary spectra, survey non-Bloch van Hove singularities, study disorder-induced effects, and explore tunable non-Hermitian flat band physics. Additionally, our framework allows proposing a hypothesis about quasi-one-dimensional bipartite non-Hermitian systems with flat bands, demonstrating their decoupling into Su-Schrieffer-Heeger chains and compact localized states across various models. Our work presents a powerful and comprehensive framework for understanding the intricate properties of non-Hermitian multiband systems, offering insights into the evolving landscape of non-Hermitian topological physics.

cond-mat.mes-hall↗
arXiv2023

Enhanced directionality of active processes in a viscoelastic bath

Active fluctuations are known to play a significant role in the intracellular transport of passive objects. However, the effect of viscoelasticity of the environment in shaping such processes is relatively less understood. Here, with a minimal experiment using a driven colloid in a viscoelastic bath, we show that viscoelasticity significantly increases the mean injected power to the passive object ($\sim 50\%$ compared to a viscous medium), for the same strength of the external driving. Additionally, we observe a notable reduction in negative work fluctuations across a wide range of driving amplitudes. These findings collectively suggest an enhanced directionality in driven processes within a viscoelastic bath, which we attribute to the emergence of interactions between the colloid and the viscoelastic medium.

cond-mat.stat-mech↗
arXiv2023

Tailoring higher-order van Hove singularities in non-Hermitian interface systems via Floquet engineering

We propose a non-Hermitian (NH) interface system formed between two NH nodal line semimetals driven by optical fields as a platform for generation and tailoring of higher-order van Hove singularities (VHS). Through an analytical analysis of the density of states (DOS), we find VHS with logarithmic divergences in the Hermitian limit. Upon introducing NH terms, four exceptional rings on two sides of the Fermi line are formed. By tuning the NH parameters and the light amplitude, we find a situation when one exceptional ring crosses the Fermi line, where a saddle point appears and results in a paired VHS around the origin. In contrast, when an exceptional contour resides at the Fermi energy, the saddle points critically get destroyed and we obtain a single peak in the DOS, with power-law divergences. These higher-order divergences that appear in an NH system have a different origin than that of the higher-order VHS in Hermitian systems, where no saddle point merging is noted. Our results suggest NH interfaces to be promising avenues for exploring higher-order VHS.

cond-mat.mes-hall↗
arXiv2023

Traversable wormholes in Rastall-Rainbow Gravity

In this paper, we investigate the existence of asymptotically flat wormhole geometries within the framework of Rastall-Rainbow modified gravity, a synthesis of two distinct theoretical models: Rastall theory and the Rainbow description. Our study uncovers that, when considering specific combinations of free parameters and equations of state, the emergence of static and spherically symmetric wormholes is not feasible within a zero-tidal-force context. By considering the subset of viable solutions, we conduct a rigorous assessment of their stability through adiabatic sound velocity analysis and scrutinize their compliance with the Weak Energy Condition (WEC). In summary, our inquiry provides insights into how the interplay between Rastall parameters and Rainbow functions may alleviate violations of energy conditions in these modified gravity scenarios.

gr-qc↗
arXiv2023

High Fidelity Fast Simulation of Human in the Loop Human in the Plant (HIL-HIP) Systems

Non-linearities in simulation arise from the time variance in wireless mobile networks when integrated with human in the loop, human in the plant (HIL-HIP) physical systems under dynamic contexts, leading to simulation slowdown. Time variance is handled by deriving a series of piece wise linear time invariant simulations (PLIS) in intervals, which are then concatenated in time domain. In this paper, we conduct a formal analysis of the impact of discretizing time-varying components in wireless network-controlled HIL-HIP systems on simulation accuracy and speedup, and evaluate trade-offs with reliable guarantees. We develop an accurate simulation framework for an artificial pancreas wireless network system that controls blood glucose in Type 1 Diabetes patients with time varying properties such as physiological changes associated with psychological stress and meal patterns. PLIS approach achieves accurate simulation with greater than 2.1 times speedup than a non-linear system simulation for the given dataset.

eess.SY↗
arXiv2023

SelfDocSeg: A Self-Supervised vision-based Approach towards Document Segmentation

Document layout analysis is a known problem to the documents research community and has been vastly explored yielding a multitude of solutions ranging from text mining, and recognition to graph-based representation, visual feature extraction, etc. However, most of the existing works have ignored the crucial fact regarding the scarcity of labeled data. With growing internet connectivity to personal life, an enormous amount of documents had been available in the public domain and thus making data annotation a tedious task. We address this challenge using self-supervision and unlike, the few existing self-supervised document segmentation approaches which use text mining and textual labels, we use a complete vision-based approach in pre-training without any ground-truth label or its derivative. Instead, we generate pseudo-layouts from the document images to pre-train an image encoder to learn the document object representation and localization in a self-supervised framework before fine-tuning it with an object detection model. We show that our pipeline sets a new benchmark in this context and performs at par with the existing methods and the supervised counterparts, if not outperforms. The code is made publicly available at: https://github.com/MaitySubhajit/SelfDocSeg

cs.CV↗
arXiv2023

A Tropical Geometric Approach To Exceptional Points

Non-Hermitian systems have been widely explored in platforms ranging from photonics to electric circuits. A defining feature of non-Hermitian systems is exceptional points (EPs), where both eigenvalues and eigenvectors coalesce. Tropical geometry is an emerging field of mathematics at the interface between algebraic geometry and polyhedral geometry, with diverse applications to science. Here, we introduce and develop a unified tropical geometric framework to characterize different facets of non-Hermitian systems. We illustrate the versatility of our approach using several examples, and demonstrate that it can be used to select from a spectrum of higher-order EPs in gain and loss models, predict the skin effect in the non-Hermitian Su-Schrieffer-Heeger model, and extract universal properties in the presence of disorder in the Hatano-Nelson model. Our work puts forth a new framework for studying non-Hermitian physics and unveils a novel connection of tropical geometry to this field.

quant-ph↗
arXiv2023

Characterizing and Tuning Exceptional Points Using Newton Polygons

The study of non-Hermitian degeneracies -- called exceptional points -- has become an exciting frontier at the crossroads of optics, photonics, acoustics, and quantum physics. Here, we introduce the Newton polygon method as a general algebraic framework for characterizing and tuning exceptional points. These polygons were first described by Isaac Newton in 1676 and are conventionally used in algebraic geometry, with deep roots in various topics in modern mathematics. We have found their surprising connection to non-Hermitian physics. We propose and illustrate how the Newton polygon method can enable the prediction of higher-order exceptional points, using a recently experimentally realized optical system. Using the paradigmatic Hatano-Nelson model, we demonstrate how our Newton Polygon method can be used to predict the presence of the non-Hermitian skin effect. As further application of our framework, we show the presence of tunable exceptional points of various orders in $PT$-symmetric one-dimensional models. We further extend our method to study exceptional points in higher number of variables and demonstrate that it can reveal rich anisotropic behaviour around such degeneracies. Our work provides an analytic recipe to understand and tune exceptional physics.

cond-mat.mes-hall↗
arXiv2023

Relativistic structure of charged quark stars in energy-momentum squared gravity

Within the context of energy-momentum squared gravity (EMSG), where non-linear matter contributions appear in the gravitational action, we derive the modified TOV equations describing the hydrostatic equilibrium of charged compact stars. We adopt two different choices for the matter Lagrangian density ($\mathcal{L}_m= p$ versus $\mathcal{L}_m= -ρ$) and investigate the impact of each one on stellar structure. Furthermore, considering a charge profile where the electric charge density $ρ_{\rm ch}$ is proportional to the standard energy density $ρ$, we solve numerically the stellar structure equations in order to obtain the mass-radius diagrams for the MIT bag model equation of state (EoS). For $\mathcal{L}_m= p$ and given a specific value of $β$ (including the uncharged case when $β= 0$), the maximum-mass values increase (decrease) substantially as the gravity model parameter $α$ becomes more negative (positive). However, for uncharged configurations and considering $\mathcal{L}_m= -ρ$, our numerical results reveal that when we increase $α$ (from a negative value) the maximum mass first increases and after reaching a maximum value it starts to decrease. Remarkably, this makes it a less trivial behavior than that caused by the first choice when we take into account the presence of electric charge ($β\neq 0$).

gr-qc↗