arXiv subjects

Priyanka Singh

Publications and source records attributed to Priyanka Singh.

At least 37 records · Page 2Linked to original sources

arXiv2025

eX-NIDS: A Framework for Explainable Network Intrusion Detection Leveraging Large Language Models

This paper introduces eX-NIDS, a framework designed to enhance interpretability in flow-based Network Intrusion Detection Systems (NIDS) by leveraging Large Language Models (LLMs). In our proposed framework, flows labelled as malicious by NIDS are initially processed through a module called the Prompt Augmenter. This module extracts contextual information and Cyber Threat Intelligence (CTI)-related knowledge from these flows. This enriched, context-specific data is then integrated with an input prompt for an LLM, enabling it to generate detailed explanations and interpretations of why the flow was identified as malicious by NIDS. We compare the generated interpretations against a Basic-Prompt Explainer baseline, which does not incorporate any contextual information into the LLM's input prompt. Our framework is quantitatively evaluated using the Llama 3 and GPT-4 models, employing a novel evaluation method tailored for natural language explanations, focusing on their correctness and consistency. The results demonstrate that augmented LLMs can produce accurate and consistent explanations, serving as valuable complementary tools in NIDS to explain the classification of malicious flows. The use of augmented prompts enhances performance by over 20% compared to the Basic-Prompt Explainer.

cs.CR↗

arXiv2025

Representation Learning on Out of Distribution in Tabular Data

The open-world assumption in model development suggests that a model might lack sufficient information to adequately handle data that is entirely distinct or out of distribution (OOD). While deep learning methods have shown promising results in handling OOD data through generalization techniques, they often require specialized hardware that may not be accessible to all users. We present TCL, a lightweight yet effective solution that operates efficiently on standard CPU hardware. Our approach adapts contrastive learning principles specifically for tabular data structures, incorporating full matrix augmentation and simplified loss calculation. Through comprehensive experiments across 10 diverse datasets, we demonstrate that TCL outperforms existing models, including FT-Transformer and ResNet, particularly in classification tasks, while maintaining competitive performance in regression problems. TCL achieves these results with significantly reduced computational requirements, making it accessible to users with limited hardware capabilities. This study also provides practical guidance for detecting and evaluating OOD data through straightforward experiments and visualizations. Our findings show that TCL offers a promising balance between performance and efficiency in handling OOD prediction tasks, which is particularly beneficial for general machine learning practitioners working with computational constraints.

cs.LG↗

arXiv2025

Random Client Selection on Contrastive Federated Learning for Tabular Data

Vertical Federated Learning (VFL) has revolutionised collaborative machine learning by enabling privacy-preserving model training across multiple parties. However, it remains vulnerable to information leakage during intermediate computation sharing. While Contrastive Federated Learning (CFL) was introduced to mitigate these privacy concerns through representation learning, it still faces challenges from gradient-based attacks. This paper presents a comprehensive experimental analysis of gradient-based attacks in CFL environments and evaluates random client selection as a defensive strategy. Through extensive experimentation, we demonstrate that random client selection proves particularly effective in defending against gradient attacks in the CFL network. Our findings provide valuable insights for implementing robust security measures in contrastive federated learning systems, contributing to the development of more secure collaborative learning frameworks

cs.LG↗

arXiv2025

Continual Contrastive Learning on Tabular Data with Out of Distribution

Out-of-distribution (OOD) prediction remains a significant challenge in machine learning, particularly for tabular data where traditional methods often fail to generalize beyond their training distribution. This paper introduces Tabular Continual Contrastive Learning (TCCL), a novel framework designed to address OOD challenges in tabular data processing. TCCL integrates contrastive learning principles with continual learning mechanisms, featuring a three-component architecture: an Encoder for data transformation, a Decoder for representation learning, and a Learner Head. We evaluate TCCL against 14 baseline models, including state-of-the-art deep learning approaches and gradient-boosted decision trees (GBDT), across eight diverse tabular datasets. Our experimental results demonstrate that TCCL consistently outperforms existing methods in both classification and regression tasks on OOD data, with particular strength in handling distribution shifts. These findings suggest that TCCL represents a significant advancement in handling OOD scenarios for tabular data.

cs.LG↗

arXiv2025

Quantifying Baryonic Feedback on Warm-Hot Circumgalactic Medium in CAMELS Simulations

The baryonic physics shaping galaxy formation and evolution are complex, spanning a vast range of scales and making them challenging to model. Cosmological simulations rely on subgrid models that produce significantly different predictions. Understanding how models of stellar and active galactic nuclei (AGN) feedback affect baryon behavior across different halo masses and redshifts is essential. Using the SIMBA and IllustrisTNG suites from the Cosmology and Astrophysics with MachinE Learning Simulations (CAMELS) project, we explore the effect of parameters governing the subgrid implementation of stellar and AGN feedback. We find that while IllustrisTNG shows higher cumulative feedback energy across all halos, SIMBA demonstrates a greater spread of baryons, quantified by the closure radius and circumgalactic medium (CGM) gas fraction. This suggests that feedback in SIMBA couples more effectively to baryons and drives them more efficiently within the host halo. There is evidence that different feedback modes are highly interrelated in these subgrid models. Parameters controlling stellar feedback efficiency significantly impact AGN feedback, as seen in the suppression of black hole mass growth and delayed activation of AGN feedback to higher mass halos with increasing stellar feedback efficiency in both simulations. Additionally, AGN feedback efficiency parameters affect the CGM gas fraction at low halo masses in SIMBA, hinting at complex, non-linear interactions between AGN and SNe feedback modes. Overall, we demonstrate that stellar and AGN feedback are intimately interwoven, especially at low redshift, due to subgrid implementation, resulting in halo property effects that might initially seem counterintuitive.

astro-ph.GA↗

arXiv2024

Vocabulary Transfer for Biomedical Texts: Add Tokens if You Can Not Add Data

Working within specific NLP subdomains presents significant challenges, primarily due to a persistent deficit of data. Stringent privacy concerns and limited data accessibility often drive this shortage. Additionally, the medical domain demands high accuracy, where even marginal improvements in model performance can have profound impacts. In this study, we investigate the potential of vocabulary transfer to enhance model performance in biomedical NLP tasks. Specifically, we focus on vocabulary extension, a technique that involves expanding the target vocabulary to incorporate domain-specific biomedical terms. Our findings demonstrate that vocabulary extension, leads to measurable improvements in both downstream model performance and inference time.

cs.CL↗

arXiv2024

Towards Explainable Network Intrusion Detection using Large Language Models

Large Language Models (LLMs) have revolutionised natural language processing tasks, particularly as chat agents. However, their applicability to threat detection problems remains unclear. This paper examines the feasibility of employing LLMs as a Network Intrusion Detection System (NIDS), despite their high computational requirements, primarily for the sake of explainability. Furthermore, considerable resources have been invested in developing LLMs, and they may offer utility for NIDS. Current state-of-the-art NIDS rely on artificial benchmarking datasets, resulting in skewed performance when applied to real-world networking environments. Therefore, we compare the GPT-4 and LLama3 models against traditional architectures and transformer-based models to assess their ability to detect malicious NetFlows without depending on artificially skewed datasets, but solely on their vast pre-trained acquired knowledge. Our results reveal that, although LLMs struggle with precise attack detection, they hold significant potential for a path towards explainable NIDS. Our preliminary exploration shows that LLMs are unfit for the detection of Malicious NetFlows. Most promisingly, however, these exhibit significant potential as complementary agents in NIDS, particularly in providing explanations and aiding in threat response when integrated with Retrieval Augmented Generation (RAG) and function calling capabilities.

cs.CR↗

arXiv2024

Probing the Circum-Galactic Medium with Fast Radio Bursts: Insights from the CAMELS Simulations

Most diffuse baryons, including the circumgalactic medium (CGM) surrounding galaxies and the intergalactic medium (IGM) in the cosmic web, remain unmeasured and unconstrained. Fast Radio Bursts (FRBs) offer an unparalleled method to measure the electron dispersion measures (DMs) of ionized baryons. Their distribution can resolve the missing baryon problem, and constrain the history of feedback theorized to impart significant energy to the CGM and IGM. We analyze the Cosmology and Astrophysics in Machine Learning (CAMEL) Simulations, using three suites: IllustrisTNG, SIMBA, and Astrid, each varying 6 parameters (2 cosmological & 4 astrophysical feedback), for a total of 183 distinct simulation models. We find significantly different predictions between the fiducial models of the suites, owing to their different implementations of feedback. SIMBA exhibits the strongest feedback, leading to the smoothest distribution of baryons, reducing the sightline-to-sightline variance in DMs between z=0-1. Astrid has the weakest feedback and the largest variance. We calculate FRB CGM measurements as a function of galaxy impact parameter, with SIMBA showing the weakest DMs due to aggressive AGN feedback and Astrid the strongest. Within each suite, the largest differences are due to varying AGN feedback. IllustrisTNG shows the most sensitivity to supernova feedback, but this is due to the change in the AGN feedback strengths, demonstrating that black holes, not stars, are most capable of redistributing baryons in the IGM and CGM. We compare our statistics directly to recent observations, paving the way for the use of FRBs to constrain the physics of galaxy formation and evolution.

astro-ph.GA↗

arXiv2024

Comparison of Models for the Warm-Hot Circumgalactic Medium around Milky Way-like Galaxies

A systematic comparison of the models of the circumgalactic medium (CGM) and their observables is crucial to understanding the predictive power of the models and constraining physical processes that affect the thermodynamics of CGM. This paper compares four analytic CGM models: precipitation, isentropic, cooling flow, and baryon pasting models for the hot, volume-filling CGM phase, all assuming hydrostatic or quasi-hydrostatic equilibrium. We show that for fiducial parameters of the CGM of a Milky-Way (MW) like galaxy ($\rm M_{vir} \sim 10^{12}~M_{\odot}$ at $z\sim 0$), the thermodynamic profiles -- entropy, density, temperature, and pressure -- show most significant differences between different models at small ($r\lesssim 30$ kpc) and large scales ($r\gtrsim 100$ kpc) while converging at intermediate scales. The slope of the entropy profile, which is one of the most important differentiators between models, is $\approx 0.8$ for the precipitation and cooling flow models, while it is $\approx0.6$ and 0 for the baryon pasting and isentropic models, respectively. We make predictions for various observational quantities for an MW mass halo for the different models, including the projected Sunyaev-Zeldovich (SZ) effect, soft X-ray emission (0.5--2 keV), dispersion measure, and column densities of oxygen ions (OVI, OVII, and OVIII) observable in absorption.

astro-ph.GA↗

arXiv2023

Mixed Nonlinear Response and Transition of Nonlinearity in a Piezoelectric Membrane

Nonlinearities play a critical role in the dynamics of mechanical resonators, enhancing sensitivity and enabling signal manipulation. Understanding the parameters affecting nonlinearities is crucial for developing strategies to counteract their effects or manipulate them for improved device performance. This study investigates the impact of fabrication-induced curvature on the dynamics of zinc oxide-based piezoelectric micromachined ultrasonic transducers (PMUTs). Our experiments reveal that these devices exhibit hardening, softening, and mixed nonlinear responses, with varying initial static displacements. Notably, PMUTs with almost flat initial static displacement exhibit hardening nonlinearity, while those with a curved initial static displacement show softening nonlinearity. An exotic mixed nonlinear response is observed for intermediate static displacement. We attribute the observed nonlinear response to the interplay between static displacement-induced quadratic nonlinearity and midplane stretching-induced cubic nonlinearity. We provide a theoretical formulation for the dynamics of the devices, which explains the experimental results and highlights the nonlinear responses and their dependence on the initial static displacement. Our findings underscore the significance of nonlinearities in the dynamics of mechanical resonators and suggest ways to optimize device performance.

physics.app-ph↗

arXiv2023

Cosmology dependency of halo masses and concentrations in hydrodynamic simulations

We employ a set of Magneticum cosmological hydrodynamic simulations that span over $15$ different cosmologies, and extract masses and concentrations of all well-resolved haloes between $z=0-1$ for critical over-densities $Δ_\texttt{vir}, Δ_{200c}, Δ_{500c}, Δ_{2500c}$ and mean overdensity $Δ_{200m}.$ We provide the first mass-concentration (Mc) relation and sparsity relation (i.e. $M_{\Delta1} - M_{\Delta2}$ mass conversion) of hydrodynamic simulations that is modelled by mass, redshift and cosmological parameters $Ω_m, Ω_b, σ_8, h_0$ as a tool for observational studies. We also quantify the impact that the Mc relation scatter and the assumption of NFW density profiles have on the uncertainty of the sparsity relation. We find that converting masses with the aid of a Mc relation carries an additional fractional scatter ($\approx 4\%$) originated from deviations from the assumed NFW density profile. For this reason we provide a direct mass-mass conversion relation fit that depends on redshift and cosmological parameters. We release the package hydro\_mc, a python tool that perform all kind of conversions presented in this paper.

astro-ph.CO↗

arXiv2022

A multi-simulation study of relativistic SZ temperature scalings in galaxy clusters and groups

The Sunyaev-Zeldovich (SZ) effect is a powerful tool in modern cosmology. With future observations promising ever improving SZ measurements, the relativistic corrections to the SZ signals from galaxy groups and clusters are increasingly relevant. As such, it is important to understand the differences between three temperature measures: (a) the average relativistic SZ (rSZ) temperature, (b) the mass-weighted temperature relevant for the thermal SZ (tSZ) effect, and (c) the X-ray spectroscopic temperature. In this work, we compare these cluster temperatures, as predicted by the {\sc Bahamas} \& {\sc Macsis}, {\sc Illustris-TNG}, {\sc Magneticum}, and {\sc The Three Hundred Project} simulations. Despite the wide range of simulation parameters, we find the SZ temperatures are consistent across the simulations. We estimate a $\simeq 10\%$ level correction from rSZ to clusters with $Y\simeq10^{-4}$~Mpc$^{-2}$. Our analysis confirms a systematic offset between the three temperature measures; with the rSZ temperature $\simeq 20\%$ larger than the other measures, and diverging further at higher redshifts. We demonstrate that these measures depart from simple self-similar evolution and explore how they vary with the defined radius of haloes. We investigate how different feedback prescriptions and resolution affect the observed temperatures, and discover the SZ temperatures are rather insensitive to these details. The agreement between simulations indicates an exciting avenue for observational and theoretical exploration, determining the extent of relativistic SZ corrections. We provide multiple simulation-based fits to the scaling relations for use in future SZ modelling.

astro-ph.CO↗

arXiv2022

Detection of AI Synthesized Hindi Speech

The recent advancements in generative artificial speech models have made possible the generation of highly realistic speech signals. At first, it seems exciting to obtain these artificially synthesized signals such as speech clones or deep fakes but if left unchecked, it may lead us to digital dystopia. One of the primary focus in audio forensics is validating the authenticity of a speech. Though some solutions are proposed for English speeches but the detection of synthetic Hindi speeches have not gained much attention. Here, we propose an approach for discrimination of AI synthesized Hindi speech from an actual human speech. We have exploited the Bicoherence Phase, Bicoherence Magnitude, Mel Frequency Cepstral Coefficient (MFCC), Delta Cepstral, and Delta Square Cepstral as the discriminating features for machine learning models. Also, we extend the study to using deep neural networks for extensive experiments, specifically VGG16 and homemade CNN as the architecture models. We obtained an accuracy of 99.83% with VGG16 and 99.99% with homemade CNN models.

cs.SD↗

arXiv2021

Using Deep Learning Techniques and Inferential Speech Statistics for AI Synthesised Speech Recognition

The recent developments in technology have re-warded us with amazing audio synthesis models like TACOTRON and WAVENETS. On the other side, it poses greater threats such as speech clones and deep fakes, that may go undetected. To tackle these alarming situations, there is an urgent need to propose models that can help discriminate a synthesized speech from an actual human speech and also identify the source of such a synthesis. Here, we propose a model based on Convolutional Neural Network (CNN) and Bidirectional Recurrent Neural Network (BiRNN) that helps to achieve both the aforementioned objectives. The temporal dependencies present in AI synthesized speech are exploited using Bidirectional RNN and CNN. The model outperforms the state-of-the-art approaches by classifying the AI synthesized audio from real human speech with an error rate of 1.9% and detecting the underlying architecture with an accuracy of 97%.

cs.LG↗

arXiv2021

Explainable AI: current status and future directions

Explainable Artificial Intelligence (XAI) is an emerging area of research in the field of Artificial Intelligence (AI). XAI can explain how AI obtained a particular solution (e.g., classification or object detection) and can also answer other "wh" questions. This explainability is not possible in traditional AI. Explainability is essential for critical applications, such as defense, health care, law and order, and autonomous driving vehicles, etc, where the know-how is required for trust and transparency. A number of XAI techniques so far have been purposed for such applications. This paper provides an overview of these techniques from a multimedia (i.e., text, image, audio, and video) point of view. The advantages and shortcomings of these techniques have been discussed, and pointers to some future directions have also been provided.

cs.LG↗

arXiv2021

SSS-PRNU: Privacy-Preserving PRNU Based Camera Attribution using Shamir Secret Sharing

Photo Response Non-Uniformity(PRNU) noise has proven to be very effective tool in camera based forensics. It helps to match a photo to the device that clicked it. In today's scenario, where millions and millions of images are uploaded every hour, it is very easy to compute this unique PRNU pattern from a couple of shared images on social profiles. This endangers the privacy of the camera owner and becomes a cause of major concern for the privacy-aware society. We propose SSS-PRNU scheme that facilitates the forensic investigators to carry out their crime investigation without breaching the privacy of the people. Thus, maintaining a balance between the two. To preserve privacy, extraction of camera fingerprint and PRNU noise for a suspicious image is computed in a trusted execution environment such as ARM TrustZone. After extraction, the sensitive information of camera fingerprint and PRNU noise is distributed into multiple obfuscated shares using Shamir secret sharing(SSS) scheme. These shares are information-theoretically secure and leak no information of underlying content. The encrypted information is distributed to multiple third-part servers where correlation is computed on a share basis between the camera fingerprint and the PRNU noise. These partial correlation values are combined together to obtain the final correlation value that becomes the basis for a match decision. Transforming the computation of the correlation value in the encrypted domain and making it well suited for a distributed environment is the main contribution of the paper. Experiment results validate the feasibility of the proposed scheme that provides a secure framework for PRNU based source camera attribution. The security analysis and evaluation of computational and storage overheads are performed to analysis the practical feasibility of the scheme.

cs.CR↗

arXiv2021

SHELBRS: Location Based Recommendation Services using Switchable Homomorphic Encryption

Location-Based Recommendation Services (LBRS) has seen an unprecedented rise in its usage in recent years. LBRS facilitates a user by recommending services based on his location and past preferences. However, leveraging such services comes at a cost of compromising one's sensitive information like their shopping preferences, lodging places, food habits, recently visited places, etc. to the third-party servers. Losing such information could be crucial and threatens one's privacy. Nowadays, the privacy-aware society seeks solutions that can provide such services, with minimized risks. Recently, a few privacy-preserving recommendation services have been proposed that exploit the fully homomorphic encryption (FHE) properties to address the issue. Though, it reduced privacy risks but suffered from heavy computational overheads that ruled out their commercial applications. Here, we propose SHELBRS, a lightweight LBRS that is based on switchable homomorphic encryption (SHE), which will benefit the users as well as the service providers. A SHE exploits both the additive as well as the multiplicative homomorphic properties but with comparatively much lesser processing time as it's FHE counterpart. We evaluate the performance of our proposed scheme with the other state-of-the-art approaches without compromising security.

cs.CR↗

arXiv2021

Detection of AI-Synthesized Speech Using Cepstral & Bispectral Statistics

Digital technology has made possible unimaginable applications come true. It seems exciting to have a handful of tools for easy editing and manipulation, but it raises alarming concerns that can propagate as speech clones, duplicates, or maybe deep fakes. Validating the authenticity of a speech is one of the primary problems of digital audio forensics. We propose an approach to distinguish human speech from AI synthesized speech exploiting the Bi-spectral and Cepstral analysis. Higher-order statistics have less correlation for human speech in comparison to a synthesized speech. Also, Cepstral analysis revealed a durable power component in human speech that is missing for a synthesized speech. We integrate both these analyses and propose a machine learning model to detect AI synthesized speech.

cs.LG↗