arXiv subjects

Vikash Singh

Publications and source records attributed to Vikash Singh.

34 records · Page 2Linked to original sources

arXiv2025

Verification of Lightning Network Channel Balances with Trusted Execution Environments (TEE)

Verifying the private liquidity state of Lightning Network (LN) channels is desirable for auditors, service providers, and network participants who need assurance of financial capacity. Current methods often lack robustness against a malicious or compromised node operator. This paper introduces a methodology for the verification of LN channel balances. The core contribution is a framework that combines Trusted Execution Environments (TEEs) with Zero-Knowledge Transport Layer Security (zkTLS) to provide strong, hardware-backed guarantees. In our proposed method, the node's balance-reporting software runs within a TEE, which generates a remote attestation quote proving the software's integrity. This attestation is then served via an Application Programming Interface (API), and zkTLS is used to prove the authenticity of its delivery. We also analyze an alternative variant where the TEE signs the report directly without zkTLS, discussing the trade-offs between transport-layer verification and direct enclave signing. We further refine this by distinguishing between "Hot Proofs" (verifiable claims via TEEs) and "Cold Proofs" (on-chain settlement), and discuss critical security considerations including hardware vulnerabilities, privacy leakage to third-party APIs, and the performance overhead of enclaved operations.

cs.CR↗

arXiv2025

Channel Balance Interpolation in the Lightning Network via Machine Learning

The Bitcoin Lightning Network is a Layer 2 payment protocol that addresses Bitcoin's scalability by facilitating quick and cost effective transactions through payment channels. This research explores the feasibility of using machine learning models to interpolate channel balances within the network, which can be used for optimizing the network's pathfinding algorithms. While there has been much exploration in balance probing and multipath payment protocols, predicting channel balances using solely node and channel features remains an uncharted area. This paper evaluates the performance of several machine learning models against two heuristic baselines and investigates the predictive capabilities of various features. Our model performs favorably in experimental evaluation, outperforming by 10% against an equal split baseline where both edges are assigned half of the channel capacity.

cs.LG↗

arXiv2025

Transfer Orthology Networks

We present Transfer Orthology Networks (TRON), a novel neural network architecture designed for cross-species transfer learning. TRON leverages orthologous relationships, represented as a bipartite graph between species, to guide knowledge transfer. Specifically, we prepend a learned species conversion layer, whose weights are masked by the biadjacency matrix of this bipartite graph, to a pre-trained feedforward neural network that predicts a phenotype from gene expression data in a source species. This allows for efficient transfer of knowledge to a target species by learning a linear transformation that maps gene expression from the source to the target species' gene space. The learned weights of this conversion layer offer a potential avenue for interpreting functional orthology, providing insights into how genes across species contribute to the phenotype of interest. TRON offers a biologically grounded and interpretable approach to cross-species transfer learning, paving the way for more effective utilization of available transcriptomic data. We are in the process of collecting cross-species transcriptomic/phenotypic data to gain experimental validation of the TRON architecture.

cs.LG↗

arXiv2025

Efficient Fine-Grained GPU Performance Modeling for Distributed Deep Learning of LLM

Training Large Language Models(LLMs) is one of the most compute-intensive tasks in high-performance computing. Predicting end-to-end training time for multi-billion parameter models distributed across hundreds of GPUs remains challenging due to complex interactions between transformer components, parallelism strategies(data, model, pipeline, tensor), and multi-tier communication. Learned models require costly sampling, while analytical models often struggle with real-world network and hardware complexities. We address this by decomposing LLMs into core computational primitives and modeling them with: (1) operator-level decomposition for fine-grained analysis; (2) lightweight sampling based hardware-aware prediction models for key operations; (3) an end-to-end prediction system integrating these components across complex parallelization strategies. Crucially, our methodology has been validated on two large-scale HPC systems. Our framework achieves low average prediction errors-4.98\% on Perlmutter(A100) and 9.38\% on Vista(GH200)-for models up to 20B parameters across 128 GPUs. Importantly, it runs entirely on CPUs, enabling rapid iteration over hardware configurations and training strategies without costly on-cluster experimentation.

cs.DC↗

arXiv2025

$K^4$: Online Log Anomaly Detection Via Unsupervised Typicality Learning

Existing Log Anomaly Detection (LogAD) methods are often slow, dependent on error-prone parsing, and use unrealistic evaluation protocols. We introduce $K^4$, an unsupervised and parser-independent framework for high-performance online detection. $K^4$ transforms arbitrary log embeddings into compact four-dimensional descriptors (Precision, Recall, Density, Coverage) using efficient k-nearest neighbor (k-NN) statistics. These descriptors enable lightweight detectors to accurately score anomalies without retraining. Using a more realistic online evaluation protocol, $K^4$ sets a new state-of-the-art (AUROC: 0.995-0.999), outperforming baselines by large margins while being orders of magnitude faster, with training under 4 seconds and inference as low as 4 $μ$s.

cs.LG↗

arXiv2025

Grammars of Formal Uncertainty: When to Trust LLMs in Automated Reasoning Tasks

Large language models (LLMs) show remarkable promise for democratizing automated reasoning by generating formal specifications. However, a fundamental tension exists: LLMs are probabilistic, while formal verification demands deterministic guarantees. This paper addresses this epistemological gap by comprehensively investigating failure modes and uncertainty quantification (UQ) in LLM-generated formal artifacts. Our systematic evaluation of five frontier LLMs reveals Satisfiability Modulo Theories (SMT) based autoformalization's domain-specific impact on accuracy (from +34.8% on logical tasks to -44.5% on factual ones), with known UQ techniques like the entropy of token probabilities failing to identify these errors. We introduce a probabilistic context-free grammar (PCFG) framework to model LLM outputs, yielding a refined uncertainty taxonomy. We find uncertainty signals are task-dependent (e.g., grammar entropy for logic, AUROC>0.93). Finally, a lightweight fusion of these signals enables selective verification, drastically reducing errors (14-100%) with minimal abstention, transforming LLM-driven formalization into a reliable engineering discipline.

cs.CL↗

arXiv2024

Bayesian Binary Search

We present Bayesian Binary Search (BBS), a novel probabilistic variant of the classical binary search/bisection algorithm. BBS leverages machine learning/statistical techniques to estimate the probability density of the search space and modifies the bisection step to split based on probability density rather than the traditional midpoint, allowing for the learned distribution of the search space to guide the search algorithm. Search space density estimation can flexibly be performed using supervised probabilistic machine learning techniques (e.g., Gaussian process regression, Bayesian neural networks, quantile regression) or unsupervised learning algorithms (e.g., Gaussian mixture models, kernel density estimation (KDE), maximum likelihood estimation (MLE)). We demonstrate significant efficiency gains of using BBS on both simulated data across a variety of distributions and in a real-world binary search use case of probing channel balances in the Bitcoin Lightning Network, for which we have deployed the BBS algorithm in a production setting.

cs.LG↗

arXiv2021

Probing Kepler's hottest small planets via homogeneous search and analysis of optical secondary eclipses and phase variations

We perform a homogeneous search for and analysis of optical occultations and phase variations of the most favorable ultra-short-period (USP) ($P<1$~d) sub-Neptunes ($R_{p}<4 R_{\oplus}$) observed by $\textit{Kepler}$ and K2, with the aim of better understanding their nature. We first selected 16 $\textit{Kepler}$ and K2 USP sub-Neptunes, based on the expected occultation signal. We filtered out stellar variability in the $\textit{Kepler}$ light curves, using a sliding linear fitting and, when required, a more sophisticated approach based on Gaussian Process regression. We simultaneously modeled the primary transit, secondary eclipse, and phase variations in a Bayesian framework, by using information from previous studies and knowledge of the Gaia parallaxes. We confirm the optical secondary eclipses for Kepler-10b ($13σ$), Kepler-78b ($9.5σ$), and K2-141b ($6.9σ$), with marginal evidence for K2-312b ($2.2σ$). We report new detections for K2-106b ($3.3σ$), K2-131b (3.2$σ$), Kepler-407b ($3.0σ$), and hints for K2-229b (2.5$σ$). For all targets with the exception of K2-229b and K2-312b, we also find phase curve variations with a confidence level higher than $2σ$. Two USP planets, namely Kepler-10b and Kepler-78b, show non-negligible nightside emission. This questions the scenario of magma-ocean worlds with inefficient heat redistribution to the night-side for both planets. Due to the youth of the Kepler-78 system and the small planetary orbital separation, the planet may still retain a collisional secondary atmosphere capable of conducting heat from the day to the night side. Instead, the presence of an outgassing magma ocean on the dayside and the low high-energy irradiation of the old host star may have enabled Kepler-10b to build up and retain a recently-formed collisional secondary atmosphere.

astro-ph.EP↗

arXiv2021

Graph Representation Learning on Tissue-Specific Multi-Omics

Combining different modalities of data from human tissues has been critical in advancing biomedical research and personalised medical care. In this study, we leverage a graph embedding model (i.e VGAE) to perform link prediction on tissue-specific Gene-Gene Interaction (GGI) networks. Through ablation experiments, we prove that the combination of multiple biological modalities (i.e multi-omics) leads to powerful embeddings and better link prediction performances. Our evaluation shows that the integration of gene methylation profiles and RNA-sequencing data significantly improves the link prediction performance. Overall, the combination of RNA-sequencing and gene methylation data leads to a link prediction accuracy of 71% on GGI networks. By harnessing graph representation learning on multi-omics data, our work brings novel insights to the current literature on multi-omics integration in bioinformatics.

q-bio.GN↗

arXiv2021

GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks

While graph neural networks (GNNs) have been shown to perform well on graph-based data from a variety of fields, they suffer from a lack of transparency and accountability, which hinders trust and consequently the deployment of such models in high-stake and safety-critical scenarios. Even though recent research has investigated methods for explaining GNNs, these methods are limited to single-instance explanations, also known as local explanations. Motivated by the aim of providing global explanations, we adapt the well-known Automated Concept-based Explanation approach (Ghorbani et al., 2019) to GNN node and graph classification, and propose GCExplainer. GCExplainer is an unsupervised approach for post-hoc discovery and extraction of global concept-based explanations for GNNs, which puts the human in the loop. We demonstrate the success of our technique on five node classification datasets and two graph classification datasets, showing that we are able to discover and extract high-quality concept representations by putting the human in the loop. We achieve a maximum completeness score of 1 and an average completeness score of 0.753 across the datasets. Finally, we show that the concept-based explanations provide an improved insight into the datasets and GNN models compared to the state-of-the-art explanations produced by GNNExplainer (Ying et al., 2019).

cs.LG↗

arXiv2020

An ultra-short period rocky super-Earth orbiting the G2-star HD 80653

Ultra-short period (USP) planets are a class of exoplanets with periods shorter than one day. The origin of this sub-population of planets is still unclear, with different formation scenarios highly dependent on the composition of the USP planets. A better understanding of this class of exoplanets will, therefore, require an increase in the sample of such planets that have accurate and precise masses and radii, which also includes estimates of the level of irradiation and information about possible companions. Here we report a detailed characterization of a USP planet around the solar-type star HD 80653 $\equiv$ EP 251279430 using the K2 light curve and 108 precise radial velocities obtained with the HARPS-N spectrograph, installed on the Telescopio Nazionale Galileo. From the K2 C16 data, we found one super-Earth planet ($R_{b}=1.613\pm0.071 R_{\oplus}$) transiting the star on a short-period orbit ($P_{\rm b}=0.719573\pm0.000021$ d). From our radial velocity measurements, we constrained the mass of HD 80653 b to $M_{b}=5.60\pm0.43 M_{\oplus}$. We also detected a clear long-term trend in the radial velocity data. We derived the fundamental stellar parameters and determined a radius of $R_{\star}=1.22\pm0.01 R_{\odot}$ and mass of $M_{\star}=1.18\pm0.04 M_{\odot}$, suggesting that HD 80653, has an age of $2.7\pm1.2$ Gyr. The bulk density ($ρ_{b} = 7.4 \pm 1.1$ g cm$^{-3}$) of the planet is consistent with an Earth-like composition of rock and iron with no thick atmosphere. Our analysis of the K2 photometry also suggests hints of a shallow secondary eclipse with a depth of 8.1$\pm$3.7 ppm. Flux variations along the orbital phase are consistent with zero. The most important contribution might come from the day-side thermal emission from the surface of the planet at $T\sim3480$ K.

astro-ph.EP↗

arXiv2019

Intelligent Surveillance of World Health Organization (WHO) Integrated Disease Surveillance and Response (IDSR) Data in Cameroon Using Multivariate Cross-Correlation

As developing countries continue to face challenges associated with infectious diseases, the need to improve infrastructure to systematically collect data which can be used to understand their outbreak patterns becomes more critical. The World Health Organization (WHO) Integrated Disease Surveillance and Response (IDSR) strategy seeks to drive the systematic collection of surveillance data to strengthen district-level reporting and to translate them into public health actions. Since the analysis of this surveillance data at the central levels of government in many developing nations has traditionally not included advanced analytics, there are opportunities for the development and exploration of computational approaches that can provide proactive insights and improve general health outcomes of infectious disease outbreaks. We propose and demonstrate a multivariate time series cross-correlation analysis as a foundational step towards gaining insight on infectious disease patterns via the pairwise computation of weighted cross-correlation scores for a specified disease across different health districts using surveillance data from Cameroon. Following the computation of weighted cross-correlation scores, we apply an anomaly detection algorithm to assess how outbreak alarm patterns align in highly correlated health districts. We demonstrate how multivariate cross-correlation analysis of weekly surveillance data can provide insight into infectious disease incidence patterns in Cameroon by identifying highly correlated health districts for a given disease. We further demonstrate scenarios in which identification of highly correlated districts aligns with alarms flagged using a standard anomaly detection algorithm, hinting at the potential of end to end solutions combining anomaly detection algorithms for flagging alarms in combination with multivariate cross-correlation analysis.

stat.AP↗

arXiv2019

Quantum Graph Neural Networks

We introduce Quantum Graph Neural Networks (QGNN), a new class of quantum neural network ansatze which are tailored to represent quantum processes which have a graph structure, and are particularly suitable to be executed on distributed quantum systems over a quantum network. Along with this general class of ansatze, we introduce further specialized architectures, namely, Quantum Graph Recurrent Neural Networks (QGRNN) and Quantum Graph Convolutional Neural Networks (QGCNN). We provide four example applications of QGNNs: learning Hamiltonian dynamics of quantum systems, learning how to create multipartite entanglement in a quantum network, unsupervised learning for spectral clustering, and supervised learning for graph isomorphism classification.

quant-ph↗

arXiv2019

Towards Probabilistic Generative Models Harnessing Graph Neural Networks for Disease-Gene Prediction

Disease-gene prediction (DGP) refers to the computational challenge of predicting associations between genes and diseases. Effective solutions to the DGP problem have the potential to accelerate the therapeutic development pipeline at early stages via efficient prioritization of candidate genes for various diseases. In this work, we introduce the variational graph auto-encoder (VGAE) as a promising unsupervised approach for learning powerful latent embeddings in disease-gene networks that can be used for the DGP problem, the first approach using a generative model involving graph neural networks (GNNs). In addition to introducing the VGAE as a promising approach to the DGP problem, we further propose an extension (constrained-VGAE or C-VGAE) which adapts the learning algorithm for link prediction between two distinct node types in heterogeneous graphs. We evaluate and demonstrate the effectiveness of the VGAE on general link prediction in a disease-gene association network and the C-VGAE on disease-gene prediction in the same network, using popular random walk driven methods as baselines. While the methodology presented demonstrates potential solely based on utilizing the topology of a disease-gene association network, it can be further enhanced and explored through the integration of additional biological networks such as gene/protein interaction networks and additional biological features pertaining to the diseases and genes represented in the disease-gene association network.

cs.LG↗

arXiv2019

Synergies between space telescopes in the photometric characterization of the atmospheres of Hot Jupiters

Previous generation of instruments have the opportunity to discover thousands of extra-solar planets and more will come with the current and future planet-search missions. In order to go one step further in the characterization of exoplanets, in this paper we describe a way to compare the photometric observation of Hot Jupiters done with space telescopes such as HST, CHEOPS, TESS, PLATO, and JWST and give the first-hand characterization on their atmospheres. We analyze a set of planetary systems hosting a Hot Jupiter for which an atmospheric template is available in the literature. For each system, we simulate the transit light curves observed by different instruments, convolving the incoming spectrum with the corresponding instrumental throughput. For each instrument, we thus measure the expected transit depth and estimate the associated uncertainty. Finally, we compare the transit depths as seen by the selected instruments and we quantify the effect of the planetary atmosphere on multi-band transit photometry. We also analyze a set of simulated scenarios with different stellar magnitudes, activity levels, transit durations and atmospheric templates to find the best cases for this kind of observational approach. We find in general that current and especially future space telescopes provide enough photometric precision to detect significant differences between the transit depths at different wavelengths. In particular, we find that the chromatic effect due to the atmosphere of the Hot Jupiters is maximized at later spectral types and that the effect of stellar activity is smaller than the measurement uncertainties.

astro-ph.EP↗

arXiv2017

Replace or Retrieve Keywords In Documents at Scale

In this paper we introduce, the FlashText algorithm for replacing keywords or finding keywords in a given text. FlashText can search or replace keywords in one pass over a document. The time complexity of this algorithm is not dependent on the number of terms being searched or replaced. For a document of size N (characters) and a dictionary of M keywords, the time complexity will be O(N). This algorithm is much faster than Regex, because regex time complexity is O(MxN). It is also different from Aho Corasick Algorithm, as it doesn't match substrings. FlashText is designed to only match complete words (words with boundary characters on both sides). For an input dictionary of {Apple}, this algorithm won't match it to 'I like Pineapple'. This algorithm is also designed to go for the longest match first. For an input dictionary {Machine, Learning, Machine learning} on a string 'I like Machine learning', it will only consider the longest match, which is Machine Learning. We have made python implementation of this algorithm available as open-source on GitHub, released under the permissive MIT License.

cs.DS↗