arXiv subjects

Mostafizur Rahman

Publications and source records attributed to Mostafizur Rahman.

At least 19 recordsLinked to original sources

arXiv2026

Quadrupole and quadratic-in-spin effects in quasicircular, spinning, asymmetric binaries

Next-generation gravitational-wave detectors will require significant improvements in current theoretical waveform models, particularly in the case of asymmetric-mass binaries. Here we provide one such improvement by calculating fully relativistic finite-size effects for small mass ratios -- primarily, fluxes of energy -- including quadratic-in-secondary-spin terms, spin-induced quadrupole terms, and tidally induced quadrupole terms, for quasicircular inspirals of a small companion into a Kerr black hole. We formulate these calculations within a multiscale waveform-generation framework in self-force theory, which could be used, with an energy-balance law we derive, to develop self-contained waveform models for asymmetric binaries involving stars orbiting black holes. Our results could additionally be used to improve other families of waveform models across all mass ratios. We present results both as complete numerical data sets on a Chebyshev grid and as analytical post-Newtonian expansions (to sixth PN order relative to the leading term in each contribution to the flux).

gr-qc↗

arXiv2026

A Logic-Reuse Approach to Nibble-based Multiplier Design for Low Power Vector Computing

Vector multiplication is a fundamental operation for AI acceleration, responsible for over 85% of computational load in convolution tasks. While essential, these operations are primary drivers of area, power, and delay in modern datapath designs. Conventional multiplier architectures often force a compromise between latency and complexity: high-speed array multipliers demand significant power, whereas sequential designs offer efficiency at the cost of throughput. This paper presents a precompute-reuse nibble multiplier architecture that bridges this gap by reformulating multiplication as a structured composition of reusable nibble-level precomputed values. The proposed design treats each operand as an independent low-precision element, decomposes it into fixed-width nibbles, and generates scaled multiples of a broadcast operand using compact shift-add logic. By replacing wide lookup tables and multiway multiplexers with logic-based precomputation and regular accumulation, the architecture decouples cycle complexity from gate delay. The design completes each 8-bit multiplication in two deterministic cycles with a short critical path, scales efficiently across vector lanes, and significantly reduces area and energy consumption. RTL implementations synthesized in TSMC 28 nm technology demonstrate up to 1.69x area reduction and 1.63x power improvement over shift-add, and nearly 2.6x area and 2.7x power savings compared to LUT-based array multipliers at 128 bit scale.

cs.AR↗

arXiv2026

Probing beyond-vacuum general relativistic effects with extreme mass-ratio inspirals

We examine extreme mass-ratio inspirals (EMRIs) as probes of beyond-vacuum general relativistic effects, accounting for both astrophysical environments and scalar Gauss-Bonnet (sGB) gravity. In beyond-vacuum scenarios, the evolution of an EMRI immersed in a cold dark matter environment modifies the gravitational wave flux and introduces additional dissipative effects such as dynamical friction. In parallel, in the beyond-general relativistic settings such as in sGB gravity, the inspiraling object carries an effective scalar charge and emits scalar radiation. Both environmental and modified-gravity effects modify the flux-balance law, thereby inducing changes in the EMRI dynamics. Using a two-timescale analysis within the fixed-frequency formalism, we compute leading-order corrections to the energy fluxes for quasi-circular, equatorial orbits in static, spherically symmetric spacetimes and construct the corresponding gravitational waveforms, which are used to quantify the accumulated gravitational wave dephasing and waveform mismatch relative to the vacuum general relativistic case. We further perform the Fisher Information Matrix analysis to estimate parameter correlations and the ability of future space-based detectors such as the Laser Interferometer Space Antenna (LISA) to disentangle environmental and modified gravity effects. Our results show that both dark matter and scalar field effects can leave measurable imprints on EMRI waveforms and that a consistent beyond-vacuum treatment is essential for robust tests of gravity.

gr-qc↗

arXiv2025

Hardware-Aware Data and Instruction Mapping for AI Tasks: Balancing Parallelism, I/O and Memory Tradeoffs

We introduce a mapping framework for deep learning inference that takes advantage of predictable neural network behavior to plan both computation and communication ahead of time. The framework generates a unified stream of instructions and data, enabling the hardware to execute operations and route information on its own, without frequent involvement from the host and with minimal off-chip memory use. This naturally reduces reliance on I/O, off-chip memory, and host control. By leveraging fine-grained message passing on a programmable, message-based compute architecture, the framework keeps data movement local and coordinates computation across the array using techniques such as stationary-weight reuse, in-array multicasting, and staged reductions. Applied to VGG-19, the framework sustains high utilization (88 to 92 percent), with over 97 percent of messages generated internally and nearly 89 percent of time consumed on-chip transfers. Computation throughput scales beyond 1 TFLOP/s on larger arrays, while traffic reductions from reuse and local aggregation reach up to 100 MB per layer. Overall, the results highlight the effectiveness of streaming-based computation and show how our mapper enables this execution style by tightly coordinating data and instruction flow across the hardware.

cs.AR↗

arXiv2025

Demystifying the 7-D Convolution Loop Nest for Data and Instruction Streaming in Reconfigurable AI Accelerators

Convolution remains the most compute-intensive operation in AI acceleration, often constituting over 80-90% of the workload. Existing approaches in spatial architectures such as coarse-grained reconfigurable arrays (CGRAs) and field-programmable gate arrays (FPGAs) frequently rely on loop unrolling or GEMM-based matrix transformations, introducing significant overhead in both data movement and instruction control. This paper presents a new framework designed to systematically demystify the 7-dimensional convolution loop nest by reinterpreting it as a hardware-centric data and instruction streaming problem. Instead of treating the loop nest as a fixed computational construct, our approach exposes its structure as a set of spatial and temporal mappings governed by hardware parameters such as compute element distribution, interconnect topology, and reconfigurability. This abstraction supports lightweight, flexible deployment of convolution without reliance on heavyweight transformations or reordering schemes. We demonstrate the application of our approach on the MAVeC accelerator. We detail the implementation of convolution operations in MAVeC and extend the framework to support full model execution on VGG-16. Our profiling reveals high PE utilization (over 90%), significant fold reuse, and scalable throughput up to 1.56 TFLOPs/sec and 12.7 KIPS for end-to-end VGG-16 inference. These results validate the efficacy of our approach in minimizing control overhead, improving data locality, and enabling efficient large-scale convolution execution without reliance on conventional transformation-based methods.

cs.AR↗

arXiv2025

Post-adiabatic waveforms from extreme mass ratio inspirals in the presence of dark matter

Extreme mass-ratio inspirals (EMRIs), in which a solar mass compact object is whirling around a supermassive black hole, act as precise tracers of the spacetime geometry and astrophysical environment around the supermassive black hole. These systems are highly sensitive to even the smallest deviations from the vacuum general relativity scenario. However, detecting these signals requires highly accurate waveform modeling up to the first post-adiabatic order, incorporating self-force effects, system parameters, and environmental influences. In this paper, we focus on the impact of dark matter on gravitational waveforms. Cold dark matter in galactic centers can be redistributed by the gravitational pull of a supermassive black hole, forming a dense, spike-like profile. When an EMRI evolves in such an environment, the interaction between the binary and the surrounding dark matter can leave distinctive imprints on the emitted waveform, and thus offer a novel way to probe the nature and distribution of dark matter. We specifically examine how dark matter modifies the background spacetime. By treating these modifications perturbatively, we present a framework to incorporate dark matter environmental effects into gravitational waveform modeling at the first post-adiabatic order.

gr-qc↗

arXiv2025

Revisiting R: Statistical Envelope Analysis for Lightweight RF Modulation Classification

Modulation classification plays a crucial role in wireless communication systems, enabling applications such as cognitive radio, spectrum monitoring, and electronic warfare. Conventional techniques often involve deep learning or complex feature extraction, which, while effective, require substantial computational resources and memory. An early approach by Chan and Gadbois in 1985 introduced a theoretical method for modulation classification using a mathematically derived parameter called R. The authors proved that the R value - the ratio of the variance to the square of the mean of the signal envelope - can be a distinguishing feature for classification. In this work, we revisit the R value and show that classification accuracy can be improved further through statistical methods. We extend R-value analysis to demonstrate its effectiveness even after signals are transformed using the Hilbert transform followed by the Short-Time Fourier Transform (STFT). Our analysis includes testing on 300000 signals across AM, DSB, and SSB classes, with each class having 100000 random variations. On average, we achieve 98.60, 97.30, and 97.90 percent classification accuracy for AM, DSB, and SSB signals after applying the Hilbert transform. Similar or better accuracies are observed after applying the STFT, reaching 98.80, 99.10, and 99.00 percent, respectively, for AM, DSB, and SSB types.

eess.SP↗

arXiv2025

PyGen: A Collaborative Human-AI Approach to Python Package Creation

The principles of automation and innovation serve as foundational elements for advancement in contemporary science and technology. Here, we introduce Pygen, an automation platform designed to empower researchers, technologists, and hobbyists to bring abstract ideas to life as core, usable software tools written in Python. Pygen leverages the immense power of autoregressive large language models to augment human creativity during the ideation, iteration, and innovation process. By combining state-of-the-art language models with open-source code generation technologies, Pygen has significantly reduced the manual overhead of tool development. From a user prompt, Pygen automatically generates Python packages for a complete workflow from concept to package generation and documentation. The findings of our work show that Pygen considerably enhances the researcher's productivity by enabling the creation of resilient, modular, and well-documented packages for various specialized purposes. We employ a prompt enhancement approach to distill the user's package description into increasingly specific and actionable. While being inherently an open-ended task, we have evaluated the generated packages and the documentation using Human Evaluation, LLM-based evaluation, and CodeBLEU, with detailed results in the results section. Furthermore, we documented our results, analyzed the limitations, and suggested strategies to alleviate them. Pygen is our vision of ethical automation, a framework that promotes inclusivity, accessibility, and collaborative development. This project marks the beginning of a large-scale effort towards creating tools where intelligent agents collaborate with humans to improve scientific and technological development substantially. Our code and generated examples are open-sourced at [https://github.com/GitsSaikat/Pygen]

cs.SE↗

arXiv2025

Guardians of the Agentic System: Preventing Many Shots Jailbreak with Agentic System

The autonomous AI agents using large language models can create undeniable values in all span of the society but they face security threats from adversaries that warrants immediate protective solutions because trust and safety issues arise. Considering the many-shot jailbreaking and deceptive alignment as some of the main advanced attacks, that cannot be mitigated by the static guardrails used during the supervised training, points out a crucial research priority for real world robustness. The combination of static guardrails in dynamic multi-agent system fails to defend against those attacks. We intend to enhance security for LLM-based agents through the development of new evaluation frameworks which identify and counter threats for safe operational deployment. Our work uses three examination methods to detect rogue agents through a Reverse Turing Test and analyze deceptive alignment through multi-agent simulations and develops an anti-jailbreaking system by testing it with GEMINI 1.5 pro and llama-3.3-70B, deepseek r1 models using tool-mediated adversarial scenarios. The detection capabilities are strong such as 94\% accuracy for GEMINI 1.5 pro yet the system suffers persistent vulnerabilities when under long attacks as prompt length increases attack success rates (ASR) and diversity metrics become ineffective in prediction while revealing multiple complex system faults. The findings demonstrate the necessity of adopting flexible security systems based on active monitoring that can be performed by the agents themselves together with adaptable interventions by system admin as the current models can create vulnerabilities that can lead to the unreliable and vulnerable system. So, in our work, we try to address such situations and propose a comprehensive framework to counteract the security issues.

cs.CR↗

arXiv2025

QuXAI: Explainers for Hybrid Quantum Machine Learning Models

The emergence of hybrid quantum-classical machine learning (HQML) models opens new horizons of computational intelligence but their fundamental complexity frequently leads to black box behavior that undermines transparency and reliability in their application. Although XAI for quantum systems still in its infancy, a major research gap is evident in robust global and local explainability approaches that are designed for HQML architectures that employ quantized feature encoding followed by classical learning. The gap is the focus of this work, which introduces QuXAI, an framework based upon Q-MEDLEY, an explainer for explaining feature importance in these hybrid systems. Our model entails the creation of HQML models incorporating quantum feature maps, the use of Q-MEDLEY, which combines feature based inferences, preserving the quantum transformation stage and visualizing the resulting attributions. Our result shows that Q-MEDLEY delineates influential classical aspects in HQML models, as well as separates their noise, and competes well against established XAI techniques in classical validation settings. Ablation studies more significantly expose the virtues of the composite structure used in Q-MEDLEY. The implications of this work are critically important, as it provides a route to improve the interpretability and reliability of HQML models, thus promoting greater confidence and being able to engage in safer and more responsible use of quantum-enhanced AI technology. Our code and experiments are open-sourced at: https://github.com/GitsSaikat/QuXAI

cs.LG↗

arXiv2024

Accelerating PageRank Algorithmic Tasks with a new Programmable Hardware Architecture

Addressing the growing demands of artificial intelligence (AI) and data analytics requires new computing approaches. In this paper, we propose a reconfigurable hardware accelerator designed specifically for AI and data-intensive applications. Our architecture features a messaging-based intelligent computing scheme that allows for dynamic programming at runtime using a minimal instruction set. To assess our hardware's effectiveness, we conducted a case study in TSMC 28nm technology node. The simulation-based study involved analyzing a protein network using the computationally demanding PageRank algorithm. The results demonstrate that our hardware can analyze a 5,000-node protein network in just 213.6 milliseconds over 100 iterations. These outcomes signify the potential of our design to achieve cutting-edge performance in next-generation AI applications.

cs.AR↗

arXiv2024

Messaging-based Adaptive Vector Computing (MAVeC) Accelerator for AI Workloads

The performance of AI accelerators is increasingly limited by data movement, memory access, and orchestration overheads rather than raw compute capability. This paper presents MAVeC, a messaging-based adaptive vector computing accelerator designed to support streaming execution and runtime configurability for AI workloads. MAVeC replaces centralized control with a message-driven execution model in which data and control propagate together across distributed hardware elements, enabling autonomous execution, flexible routing, and efficient coordination. We validate MAVeC's core hardware constructs and execution model using matrix multiplication and convolution workloads under a cycle-accurate, system-level ASIC design in TSMC 28 nm, capturing computation, communication, and reduction. MAVeC sustains greater than 97 percent array utilization across hardware scales and problem sizes by translating spatial capacity into effective computation. Once inputs are brought in, over 90 percent of communication remains on-chip through coordinated temporal reuse, spatial multicast, and on-fabric partial-sum reduction. On a 64x64 SiteO array, MAVeC sustains over 5 TFLOPs per second while reducing end-to-end latency. Compared to TPU-style systolic arrays and MEISSA under compute-centric models, MAVeC achieves 1.5-2x lower latency. When evaluated against optimized NVIDIA H100 FP32 kernels, MAVeC sustains 5.8-6.1 TFLOPs per second, delivering a consistent 6.0-7.2x throughput advantage across problem sizes. Energy results show that MAVeC converts higher instantaneous power into lower total energy by shortening execution time and amortizing data movement. These results demonstrate that message-driven execution provides an effective architectural foundation for overcoming data movement and orchestration bottlenecks, enabling scalable, high-utilization accelerators for future AI workloads.

cs.AR↗

arXiv2023

Perturbing the perturbed: Stability of quasinormal modes in presence of a positive cosmological constant

In this work, we wish to address the question -- whether the quasi-normal modes, the characteristic frequencies associated with perturbed black hole spacetimes, central to the stability of these black holes, are themselves stable. Though the differential operator governing the perturbation of black hole spacetimes is self-adjoint, the boundary conditions are dissipative in nature, so that the spectral theorem becomes inapplicable, and there is no guarantee regarding the stability of the quasi-normal modes. We have provided a general method of transforming to the hyperboloidal coordinate system, for both asymptotically flat and asymptotically de Sitter spacetimes which neatly captures the dissipative boundary conditions, and the differential operator becomes non-self-adjoint. Employing the pseudospectrum analysis and numerically implementing the same through Chebyshev's spectral method, we present how the quasi-normal modes will drift away from their unperturbed values under external perturbation of the scattering potential. Intriguingly, for strong enough perturbation, even the fundamental quasi-normal mode, associated with gravitational perturbations, drifts away from its unperturbed position for asymptotically de Sitter black holes, in stark contrast to the case of asymptotically flat black holes. Besides presenting several other interesting results, specifically for asymptotically de Sitter black holes, we also discuss the implications of the instability of the fundamental quasi-normal mode on the strong cosmic censorship conjecture.

gr-qc↗

arXiv2023

Traversable wormholes in bi-metric gravity

The ghost-free bi-metric gravity theory is a viable theory of gravity that explores the interaction between a massless and a massive graviton and can be described in terms of two dynamical metrics. In this paper, we present an exact static, spherically symmetric vacuum solution within this theory. The solution is spatially Schwarzschild-de Sitter, with the value of the cosmological constant determined by the graviton mass and the interaction parameters of the theory. Notably, for specific parameter ranges, the solution represents a traversable Lorentzian wormhole that violates the weak energy condition near its throat. Furthermore, we have investigated the evolution of scalar and electromagnetic fields in this wormhole spacetime and observed the presence of arbitrarily long-lived quasi-resonant modes in the quasinormal spectrum.

gr-qc↗

arXiv2023

Probing astrophysical environment with eccentric extreme mass-ratio inspirals

The discovery of gravitational waves and black holes has started a new era of gravitational wave astronomy that allows us to probe the underpinning features of gravity and astrophysics in extreme environments of the universe. In this article, we investigate one such study with an extreme mass-ratio inspiral system where the primary object is a spherically symmetric static black hole immersed in a dark matter halo governed by the Hernquist density distribution. We consider the eccentric equatorial orbital motion of the steller-mass object orbiting around the primary and compute measurable effects. We examine the behaviour of dark matter mass and halo radius in generated gravitational wave fluxes and the evolution of eccentric orbital parameters -- eccentricity and semi-latus rectum. We further provide an estimate of gravitational wave dephasing and find the seminal role of low-frequency detectors in the observational prospects of such an astrophysical environment.

gr-qc↗

arXiv2023

Regularized Stable Kerr Black Hole: Cosmic Censorships, Shadow and Quasi-Normal Modes

Black hole solutions in general relativity come with pathologies such as singularity and mass inflation instability, which are believed to be cured by a yet-to-be-found quantum theory of gravity. Without such consistent description, one may model theory-agnostic phenomenological black holes that bypass the aforesaid issues. These so-called regular black holes are extensively studied in the literature using parameterized modifications over the black hole solutions of general relativity. However, since there exist several ways to model such black holes, it is important to study the consistency and viability of these solutions from both theoretical and observational perspectives. In this work, we consider a recently proposed model of regularized stable rotating black holes having two extra parameters in addition to the mass and spin of a Kerr solution. We start by computing their quasi-normal modes under scalar perturbation and investigate the impact of those additional parameters on black hole stability. In the second part, we study the shadow structures of these regularized black holes and obtain stringent bounds on the parameter space requiring consistency with Event Horizon Telescope observations of $M87^*$ and $Sgr\, A^*$ shadows.

gr-qc↗

arXiv2022

Prospects for determining the nature of the secondaries of extreme mass-ratio inspirals using the spin-induced quadrupole deformation

The measurement of multipole moments of astrophysical objects through gravitational wave (GW) observations provides a novel way to distinguish black holes from other astrophysical objects. This paper studies the gravitational wave radiation from an extreme mass ratio inspiral (EMRI) system consisting of a supermassive Kerr black hole (the primary object) and a spinning stellar-mass compact object (the secondary object). The quadrupolar deformation induced by the spin of the secondary is different for different astrophysical objects. We compute the effect of the quadrupolar deformation on the GW phase and provide an order of magnitude estimate of whether LISA can distinguish different astrophysical objects through GW phase measurement. We find that although LISA can not distinguish between a black hole and a neutron star, it can distinguish black holes from a large variety of highly spinning astrophysical objects like superspinars and highly deformable exotic compact objects like boson stars for EMRI systems with relatively large mass ratio ($q\sim 10^{-4}$). Furthermore, we show that the effect of spin-induced quadrupolar deformation on the GW phase for white dwarf and brown dwarf-EMRI systems can be quite significant even for small values of mass ratio ($q\lesssim 10^{-6}$).

gr-qc↗

arXiv2022

Echoes from braneworld wormholes

We have studied the stability of wormhole geometries, under massless scalar, electromagnetic and axial gravitational perturbations, in the context of higher dimensional spacetimes. Intriguingly, the construction of a wormhole spacetime in the presence of higher dimensions, known as braneworld wormholes, \emph{does not} require the existence of exotic matter fields, unlike the scenario in four spacetime dimensions. Being a nonvacuum spacetime, the effective potential experienced by the axial gravitational perturbation differs considerably from the scenarios involving black holes. In particular, the present work provides one of the first attempts to study the gravitational perturbations of the wormhole spacetimes. Our analysis, involving both analytical and numerical techniques, demonstrates that there are echoes in the time domain signal of all the perturbations and the echo time delay is intimately related to the parameters originating from higher dimensions. Thereby combining the attempt to search for wormholes and extra dimensions, with the existence of gravitational wave echoes. Implications and future directions have also been discussed.

gr-qc↗