arXiv subjects

Chayan Chatterjee

Publications and source records attributed to Chayan Chatterjee.

18 recordsLinked to original sources

arXiv2026

A Fast and Scalable Transformer Pipeline for Binary Black Hole Detection

With the projected increase in the detection rate of compact-binary coalescences in the coming decade, there is critical need to develop fast, robust, and scalable alternatives to matched filtering for gravitational-wave searches. Transformer models have revolutionized natural language and audio processing but their application to gravitational-wave astronomy is still largely unexplored. In this work, we introduce \castor, a transformer-based coincident search pipeline for detecting binary black hole gravitational-wave signals from Advanced LIGO detectors. One of the major features of our model is that it allows the false-alarm rate to be estimated via time slides cheaply without requiring repeated evaluations of the neural network. We evaluate \castor\ on datasets from the Machine-Learning Gravitational-Wave Search Challenge (MLGWSC-1) and on approximately five months of real O3b observing strain. When tested on benchmark datasets, \castor\ ranks among the most sensitive machine-learning pipelines and successfully recovers the majority of confident events from the GWTC-3 catalog that lie within its training range. We also benchmark \castor\ against another transformer architecture, GW-Whisper, a domain-adaptation of OpenAI's audio foundation model. We find that \castor\ substantially outperforms the repurposed audio model in sensitivity and also reduces the computational cost of background estimation by a factor of 20. Our results demonstrate a highly practical, scalable approach for deep-learning gravitational wave searches and empirical background estimation for future observing runs.

gr-qc↗

arXiv2026

Rapid data quality investigations of gravitational-wave events with the Data Quality Report Builder toolkit

We present the Data Quality Report Builder toolkit, DQRbuild, a suite of data quality tools that have been developed to vet gravitational-wave events in preparation for the fourth LIGO-Virgo-KAGRA observing run. We explain the main functionality and the many scientific tests that we support. To validate the performance of the tools included in the toolkit, we run a series of tests on all significant candidates shared as public alerts in the third observing run to compare against what was manually reported using human intervention. We find that these automated tools can now identify 96% of the problems identified by humans during this previous observing run, with a 24% false alarm rate. We conclude with a commentary on the prospects and potential challenges for fully automating the process of vetting the data quality for gravitational-wave events identified in future observing runs.

astro-ph.IM↗

arXiv2026

ArchGEM: an Advanced Data Analysis Tool for Analyzing Scattered Light Noise in LIGO

Scattered light is one of the most common sources of non-stationary noise at low frequencies in Advanced LIGO detectors. It appears as arch-like features in time-frequency spectrograms, produced when stray light reflects from moving surfaces and recombines with the main interferometer beam. In this study, we present ArchGEM, an automated framework for identifying and characterizing these arches and recovering the physical properties of the scattering surfaces. ArchGEM combines a prominence-based peak-finding method with a Gaussian Mixture Model clustering approach to capture a range of scattered-light morphologies across different detector conditions. We apply ArchGEM to scattered light glitches across Advanced LIGO observing runs O3 (2019--2020) and O4 (2023--2024). We find that the average frequency distributions of this noise span 15--25 Hz in O3a and O4, but increase to 20--40 Hz during O3b. Typical inferred surface velocities are 0.2--0.5 $\mu$m/s, and inferred surface displacements are 0.1--0.3 $\mu$m. The Gaussian Mixture Model performs most consistently for complex or overlapping features, with mean frequency offsets within 5 Hz of the Gravity Spy baseline. Our results show that ArchGEM provides a practical tool for detector characterization by linking observed spectrogram features to the motion of scattering surfaces and helping guide future mitigation of scattered light noise in current and next-generation interferometers. By quantifying the temporal and spectral behavior of scattered light, ArchGEM provides a robust framework for diagnosing noise sources and guiding targeted mitigation strategies in future detector upgrades.

astro-ph.IM↗

arXiv2026

Interpretable Analytic Formulae for GWTC-4 Binary Black Hole Population Properties via Symbolic Regression

Recent LIGO-Virgo-KAGRA (LVK) analyses have revealed complex structure in the binary black hole (BBH) population, including distinct features in the primary mass spectrum and nontrivial spin-mass correlations. However, the phenomenological models used to capture these features often lack analytic transparency, making it difficult to isolate robust physical laws from modeling artifacts. To address this, symbolic regression is applied to the posterior inference products of the GWTC-4 catalog, discovering compact, closed-form analytic expressions for four key population relationships: (i) the merger-rate evolution with redshift; (ii) the mass-ratio dependence of the effective-spin distribution; (iii) the redshift evolution of the effective-spin distribution; and (iv) the conditional mass-ratio distributions associated with the 10 solar mass and 35 solar mass primary mass peaks. This framework successfully compresses both rigid and highly flexible models into differentiable phenomenological laws, dynamically recovering a consistent low-redshift merger-rate slope without assuming an a priori power-law form. The exact analytic derivatives provided by symbolic regression show that the mass ratio--effective spin and redshift--effective spin correlations are robustly driven by broadening of the posterior widths rather than shifts in the mean. Furthermore, qualitatively distinct functional forms for the mass-ratio distributions conditioned on the 10 solar mass and 35 solar mass primary mass peaks are identified. These closed-form expressions enable exact analytic gradient diagnostics and compact surrogate summaries, particularly for flexible numerical posteriors that are not otherwise available in low-dimensional analytic form. They also facilitate rapid downstream calculations for rate forecasting, formation channel comparison, and stochastic background estimation.

astro-ph.CO↗

arXiv2026

The Sound of Noise: Leveraging the Inductive Bias of Pre-trained Audio Transformers for Glitch Identification in LIGO

Transient noise artifacts, or glitches, fundamentally limit the sensitivity of gravitational-wave (GW) interferometers and can mimic true astrophysical signals, particularly the short-duration intermediate-mass black hole (IMBH) mergers. Current glitch classification methods, such as Gravity Spy, rely on supervised models trained from scratch using labeled datasets. These approaches suffer from a significant ``label bottleneck," requiring massive, expertly annotated datasets to achieve high accuracy and often struggling to generalize to new glitch morphologies or exotic GW signals encountered in observing runs. In this work, we present a novel cross-domain framework that treats GW strain data through the lens of audio processing. We utilize the Audio Spectrogram Transformer (AST), a model pre-trained on large-scale audio datasets, and adapt it to the GW domain. Instead of learning time-frequency features from scratch, our method exploits the strong inductive bias inherent in pre-trained audio models, transferring learned representations of natural sound to the characterization of detector noise and GW signals, including IMBHs. We validate this approach by analyzing strain data from the third (O3) and fourth (O4) observing runs of the LIGO detectors. We used t-Distributed Stochastic Neighbor Embedding (t-SNE), an unsupervised clustering technique, to visualize the AST-derived embeddings of signals and glitches, revealing well-separated groups that align closely with independently validated Gravity Spy glitch classes. Our results indicate that the inductive bias from audio pre-training allows superior feature extraction compared to traditional supervised techniques, offering a robust, data-efficient pathway for discovering new, anomalous transients, and classifying complex noise artifacts in the era of next-generation detectors.

astro-ph.IM↗

arXiv2025

Machine Learning Confirms GW231123 is a "Lite" Intermediate Mass Black Hole Merger

The LIGO-Virgo-KAGRA Collaboration recently reported GW231123, a black hole merger with total mass of around 190-265 solar mass. This event adds to the growing evidence of "lite" intermediate mass black hole (IMBH) discoveries of post-merger black holes >100 solar mass. GW231123 posed several data analysis challenges owing to waveform-model systematics and presence of noise artifacts called glitches. We present the first comprehensive machine learning analysis to further validate this event, strengthen its astrophysical inference, and characterize instrumental noise in its vicinity. Our approach uses a combination of tools tailored for specific analyses: GW-Whisper, an adaptation of OpenAI's audio transformer, ArchGEM, a Gaussian mixture model-based soft clustering and density approximation software and AWaRe, a convolutional autoencoder. We identify the data segment containing the merger with >70% confidence in both detectors and verify its astrophysical origin. We then characterize the scattered light glitch around the event, providing the first physically interpretable parameters for the glitch. We also reconstruct the real waveforms from the data with slightly better agreement to model-agnostic reconstructions than to quasi-circular models, hinting at possible astrophysics beyond current waveform families (such as non-circular orbits or environmental imprints). Finally, by demonstrating high-fidelity waveform reconstructions for simulated mergers with total masses between 100-1000 solar mass, we show that our method can confidently probe the IMBH regime. Our integrated framework offers a powerful complementary tool to traditional pipelines for rapid, robust analysis of massive, glitch-contaminated events.

astro-ph.HE↗

arXiv2024

Pre-trained Audio Transformer as a Foundational AI Tool for Gravitational Waves

As gravitational wave detectors become more advanced and sensitive, the number of signals recorded by Advanced LIGO and Virgo from merging compact objects is expected to rise dramatically. This surge in detection rates necessitates the development of adaptable, scalable, and efficient tools capable of addressing a wide range of tasks in gravitational wave astronomy. Foundational AI models present a transformative opportunity in this context by providing a unified framework that can be fine tuned for diverse applications while leveraging the power of large scale pre training. In this work, we explore how advanced transformer models, specifically Whisper by OpenAI, can be adapted as a foundational model for gravitational wave data analysis. By fine tuning the encoder model of Whisper, originally trained on extensive audio data, and combining it with neural networks for specialized tasks, we achieve reliable results in detecting astrophysical signals and classifying transient noise artifacts or glitches. This represents the first application of open source transformer models, pre trained on unrelated tasks, for gravitational wave research, demonstrating their potential to enable versatile and efficient data analysis in the era of rapidly increasing detection rates.

gr-qc↗

arXiv2024

No Glitch in the Matrix: Robust Reconstruction of Gravitational Wave Signals Under Noise Artifacts

Gravitational wave observations by ground based detectors such as LIGO and Virgo have transformed astrophysics, enabling the study of compact binary systems and their mergers. However, transient noise artifacts, or glitches, pose a significant challenge, often obscuring or mimicking signals and complicating their analysis. In this work, we extend the Attention-boosted Waveform Reconstruction network to address glitch mitigation, demonstrating its robustness in reconstructing waveforms in the presence of real glitches from the third observing run of LIGO. Without requiring explicit training on glitches, AWaRe accurately isolates gravitational wave signals from data contaminated by glitches spanning a wide range of amplitudes and morphologies. We evaluate this capability by investigating the events GW191109 and GW200129, which exhibit strong evidence of anti-aligned spins and spin precession respectively, but may be adversely affected by data quality issues. We find that, regardless of the potential presence of glitches in the data, AWaRe reconstructs both waveforms with high accuracy. Additionally, we perform a systematic study of the performance of AWaRe on a simulated catalog of injected waveforms in real LIGO glitches and obtain reliable reconstructions of the waveforms. By subtracting the AWaRe reconstructions from the data, we show that the resulting residuals closely align with the background noise that the waveforms were injected in. The robustness of AWaRe in mitigating glitches, despite being trained exclusively on GW signals and not explicitly on glitches, highlights its potential as a powerful tool for improving the reliability of searches and characterizing noise artifacts.

gr-qc↗

arXiv2024

Navigating Unknowns: Deep Learning Robustness for Gravitational Wave Signal Reconstruction

We present a rapid and reliable deep learning-based method for gravitational wave signal reconstruction from elusive, generic binary black hole mergers in LIGO data. We demonstrate that our model, \texttt{AWaRe}, effectively recovers gravitational waves with parameters it has not encountered during training. This includes features like higher black hole masses, additional harmonics, eccentricity, and varied waveform systematics, which introduce complex modulations in the waveform's amplitudes and phases. The accurate reconstructions of these unseen signal characteristics demonstrates \texttt{AWaRe}'s ability to handle complex features in the waveforms. By directly incorporating waveform reconstruction uncertainty estimation into the \texttt{AWaRe} framework, we show that for real gravitational wave events, the uncertainties in \texttt{AWaRe}'s reconstructions align closely with those achieved by benchmark algorithms like BayesWave and coherent WaveBurst. The robustness of our model to real gravitational wave events and its ability to extrapolate to unseen data open new avenues for investigations in various aspects of gravitational wave astrophysics and data analysis, including tests of General Relativity and the enhancement of current gravitational wave search methodologies.

gr-qc↗

arXiv2024

Reconstruction of binary black hole harmonics in LIGO using deep learning

Gravitational wave signals from coalescing compact binaries in the LIGO and Virgo interferometers are primarily detected by the template based matched filtering method. While this method is optimal for stationary and Gaussian data scenarios, its sensitivity is often affected by non stationary noise transients in the detectors. Moreover, most of the current searches do not account for the effects of precession of black hole spins and higher order waveform harmonics, focusing solely on the leading order quadrupolar modes. This limitation impacts our search for interesting astrophysical sources, such as intermediate mass black hole binaries and hierarchical mergers. Here we show for the first time that deep learning can be used for accurate waveform reconstruction of precessing binary black hole signals with higher order modes. This approach can also be adapted into a rapid trigger generation algorithm to enhance online searches. Our model, tested on simulated injections in real LIGO noise from the third observing run achieved high-degree of overlap with injected signals. This accuracy was consistent across a wide range of black hole masses and spin configurations chosen for this study. When applied to real gravitational wave events, our reconstructions achieved between 0.85 and 0.98 overlaps with those obtained by Coherent WaveBurst (unmodeled) and LALInference (modeled) analyses. These results suggest that deep learning is a potent tool for analyzing signals from a diverse catalog of compact binaries.

gr-qc↗

arXiv2023

Applications of Deep Learning to physics workflows

Modern large-scale physics experiments create datasets with sizes and streaming rates that can exceed those from industry leaders such as Google Cloud and Netflix. Fully processing these datasets requires both sufficient compute power and efficient workflows. Recent advances in Machine Learning (ML) and Artificial Intelligence (AI) can either improve or replace existing domain-specific algorithms to increase workflow efficiency. Not only can these algorithms improve the physics performance of current algorithms, but they can often be executed more quickly, especially when run on coprocessors such as GPUs or FPGAs. In the winter of 2023, MIT hosted the Accelerating Physics with ML at MIT workshop, which brought together researchers from gravitational-wave physics, multi-messenger astrophysics, and particle physics to discuss and share current efforts to integrate ML tools into their workflows. The following white paper highlights examples of algorithms and computing frameworks discussed during this workshop and summarizes the expected computing needs for the immediate future of the involved fields.

hep-ex↗

arXiv2022

Pre-merger sky localization of gravitational waves from binary neutron star mergers using deep learning

The simultaneous observation of gravitational waves (GW) and prompt electromagnetic counterparts from the merger of two neutron stars can help reveal the properties of extreme matter and gravity during and immediately after the final plunge. Rapid sky localization of these sources is crucial to facilitate such multi-messenger observations. Since GWs from binary neutron star (BNS) mergers can spend up to 10-15 mins in the frequency bands of the detectors at design sensitivity, early warning alerts and pre-merger sky localization can be achieved for sufficiently bright sources, as demonstrated in recent studies. In this work, we present pre-merger BNS sky localization results using CBC-SkyNet, a deep learning model capable of inferring sky location posterior distributions of GW sources at orders of magnitude faster speeds than standard Markov Chain Monte Carlo methods. We test our model's performance on a catalog of simulated injections from Sachdev et al. (2020), recovered at 0-60 secs before merger, and obtain comparable sky localization areas to the rapid localization tool BAYESTAR. These results show the feasibility of our model for rapid pre-merger sky localization and the possibility of follow-up observations for precursor emissions from BNS mergers.

astro-ph.HE↗

arXiv2022

Rapid localization of gravitational wave sources from compact binary coalescences using deep learning

The mergers of neutron star-neutron star and neutron star-black hole binaries are the most promising gravitational wave events with electromagnetic counterparts. The rapid detection, localization and simultaneous multi-messenger follow-up of these sources is of primary importance in the upcoming science runs of the LIGO-Virgo-KAGRA Collaboration. While prompt electromagnetic counterparts during binary mergers can last less than two seconds, the time scales of existing localization methods that use Bayesian techniques, varies from seconds to days. In this paper, we propose the first deep learning-based approach for rapid and accurate sky localization of all types of binary coalescences, including neutron star-neutron star and neutron star-black hole binaries for the first time. Specifically, we train and test a normalizing flow model on matched-filtering output from gravitational wave searches. Our model produces sky direction posteriors in milliseconds using a single P100 GPU, which is three to six orders of magnitude faster than Bayesian techniques.

gr-qc↗

arXiv2022

Rapid Mass Parameter Estimation of Binary Black Hole Coalescences Using Deep Learning

Deep learning can be used to drastically decrease the processing time of parameter estimation for coalescing binaries of compact objects including black holes and neutron stars detected in gravitational waves (GWs). As a first step, we present two neural network models trained to rapidly estimate the posterior distributions of the chirp mass and mass ratio of a detected binary black hole system from the GW strain data of LIGO Hanford and Livingston Observatories. Using these parameters the component masses can be predicted, which has implications for the prediction of the likelihood that a merger contains a neutron star. The results are compared to the 'gold standard' of parameter estimation of gravitational waves used by the LIGO-Virgo Collaboration (LVC), LALInference. Our models predict posterior distributions consistent with that from LALInference while using orders of magnitude less processing time once the models are trained. The median predictions are within the 90% credible intervals of LALInference for all predicted parameters when tested on real binary black hole events detected during the LVC's first and second observing runs. We argue that deep learning has strong potential for low-latency high-accuracy parameter estimation suitable for real-time GW search pipelines.

astro-ph.IM↗

arXiv2021

Extraction of Binary Black Hole Gravitational Wave Signals from Detector Data Using Deep Learning

Accurate extractions of the detected gravitational wave (GW) signal waveforms are essential to validate a detection and to probe the astrophysics behind the sources producing the GWs. This however could be difficult in realistic scenarios where the signals detected by existing GW detectors could be contaminated with non-stationary and non-Gaussian noise. While the performance of existing waveform extraction methods are optimal, they are not fast enough for online application, which is important for multi-messenger astronomy. In this paper, we demonstrate that a deep learning architecture consisting of Convolutional Neural Network and bidirectional Long Short-Term Memory components can be used to extract binary black hole (BBH) GW waveforms from realistic noise in a few milli-seconds. We have tested our network systematically on injected GW signals, with component masses uniformly distributed in the range of 10 to 80 solar masses, on Gaussian noise and LIGO detector noise. We find that our model can extract GW waveforms with overlaps of more than 0.95 with pure Numerical Relativity templates for signals with signal-to-noise ratio (SNR) greater than six, and is also robust against interfering glitches. We then apply our model to all ten detected BBH events from the first (O1) and second (O2) observation runs, obtaining greater than 0.97 overlaps for all ten extracted BBH waveforms with the corresponding pure templates. We discuss the implication of our result and its future applications to GW localization and mass estimation.

gr-qc↗

arXiv2020

Enhancing Gravitational-Wave Science with Machine Learning

Machine learning has emerged as a popular and powerful approach for solving problems in astrophysics. We review applications of machine learning techniques for the analysis of ground-based gravitational-wave detector data. Examples include techniques for improving the sensitivity of Advanced LIGO and Advanced Virgo gravitational-wave searches, methods for fast measurements of the astrophysical parameters of gravitational-wave sources, and algorithms for reduction and characterization of non-astrophysical detector noise. These applications demonstrate how machine learning techniques may be harnessed to enhance the science that is possible with current and future gravitational-wave detectors.

astro-ph.HE↗

arXiv2019

Using Deep Learning to Localize Gravitational Wave Sources

In this paper, we report on the construction of a deep Artificial Neural Network (ANN) to localize simulated gravitational wave signals in the sky with high accuracy. We have modelled the sky as a sphere and have considered cases where the sphere is divided into 18, 50, 128, 1024, 2048 and 4096 sectors. The sky direction of the gravitational wave source is estimated by classifying the signal into one of these sectors based on it's right ascension and declination values for each of these cases. In order to do this, we have injected simulated binary black hole gravitational wave signals of component masses sampled uniformly between 30-80 solar mass into Gaussian noise and used the whitened strain values to obtain the input features for training our ANN. We input features such as the delays in arrival times, phase differences and amplitude ratios at each of the three detectors Hanford, Livingston and Virgo, from the raw time-domain strain values as well as from analytical versions of these signals, obtained through Hilbert transformation. We show that our model is able to classify gravitational wave samples, not used in the training process, into their correct sectors with very high accuracy (>90%) for coarse angular resolution using 18, 50 and 128 sectors. We also test our localization on test samples with injection parameters of the published LIGO binary black hole merger events GW150914, GW170818 and GW170823 for 1024, 2048 and 4096 sectors and compare the result with that from BAYESTAR and Parameter Estimation (PE). In addition, we report that the time taken by our model to localize one GW signal is around 0.018 secs on 14 Intel Xeon CPU cores.

astro-ph.IM↗

arXiv2019

Dark Matter Self Interactions and its Impact on Large Scale Structures

The LambdaCDM model of cosmology, though very successful at large scales, has some discrepancy with observations at the galactic and sub-galactic scales. These include the core-cusp problem, missing satellites problem etc. Spergel and Steingardt (2000) proposed that if dark matter undergoes feeble self interactions with each other, then such problems can be averted. In this thesis, a two-component Feebly Interacting Massive Particle (FIMP) dark matter model involving two singlet scalar fields capable of self-interactions has been proposed and its impact on large scale structure formation has been studied through cosmological simulations. The proposed model involves simple extensions of the Standard Model with two singlet scalar fields formed non-thermally through the decay of heavier particles in the very early universe. These particles acquire their relic abundance through a freeze-in mechanism. The coupled Boltzmann equation of the FIMP-FIMP model was solved and the relic densities for different values of the coupling parameters were obtained and matched with PLANCK results. The impact of dark matter self interactions was studied through cosmological simulations using a modified version of the parallel TreePM code GADGET-2 and the halo mass function and halo catalog for different dark matter self interaction cross sections were obtained. Lastly, the newly developed Effective Theory of Structure Formation (ETHOS) framework which is a new and innovative paradigm in the study of the cosmological effects of different dark matter models was studied and using the public code, ETHOS-CAMB the signatures of dark acoustic oscillations in the matter power spectrum for a particular dark matter model was obtained.

astro-ph.CO↗