Inference-optimized AI and high performance computing for gravitational wave detection at scale

Abstract

We introduce an ensemble of artificial intelligence models for gravitational wave detection that we trained in the Summit supercomputer using 32 nodes, equivalent to 192 NVIDIA V100 GPUs, within 2 hours. Once fully trained, we optimized these models for accelerated inference using NVIDIA TensorRT. We deployed our inference-optimized AI ensemble in the ThetaGPU supercomputer at Argonne Leadership Computer Facility to conduct distributed inference. Using the entire ThetaGPU supercomputer, consisting of 20 nodes each of which has 8 NVIDIA A100 Tensor Core GPUs and 2 AMD Rome CPUs, our NVIDIA TensorRT-optimized AI ensemble processed an entire month of advanced LIGO data (including Hanford and Livingston data streams) within 50 seconds. Our inference-optimized AI ensemble retains the same sensitivity of traditional AI models, namely, it identifies all known binary black hole mergers previously identified in this advanced LIGO dataset and reports no misclassifications, while also providing a 3X inference speedup compared to traditional artificial intelligence models. We used time slides to quantify the performance of our AI ensemble to process up to 5 years worth of advanced LIGO data. In this synthetically enhanced dataset, our AI ensemble reports an average of one misclassification for every month of searched advanced LIGO data. We also present the receiver operating characteristic curve of our AI ensemble using this 5 year long advanced LIGO dataset. This approach provides the required tools to conduct accelerated, AI-driven gravitational wave detection at scale.

arXiv comment19 pages, 8 figures; v2. Accepted to Frontiers in Artificial Intelligence, Special Issue: Efficient AI in Particle Physics and Astrophysics

References in corpus (54)

• GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiral
• Advanced Virgo: a 2nd generation interferometric gravitational wave detector
• WaveNet: A Generative Model for Raw Audio
• Gravitational Waves and Gamma-rays from a Binary Neutron Star Merger: GW170817 and GRB 170817A
• GWTC-3: Compact Binary Coalescences Observed by LIGO and Virgo During the Second Part of the Third Observing Run
• Binary Black Hole Mergers in the first Advanced LIGO Observing Run
• Tests of General Relativity with Binary Black Holes from the second LIGO-Virgo Gravitational-Wave Transient Catalog
• GW150914: The Advanced LIGO Detectors in the Era of First Discoveries
• Theoretical Physics Implications of the Binary Black-Hole Mergers GW150914 and GW151226
• A mildly relativistic wide-angle outflow in the neutron star merger GW170817
• Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation: Results with Advanced LIGO Data
• Inspiral-merger-ringdown waveforms of spinning, precessing black-hole binaries in the effective-one-body formalism
• Gravity Spy: Integrating Advanced LIGO Detector Characterization, Machine Learning, and Citizen Science
• Deep Neural Networks to Enable Real-time Multimessenger Astrophysics
• The LIGO Open Science Center
• Real-time gravitational-wave science with neural posterior estimation
• Matching matched filtering with deep networks in gravitational-wave astronomy
• Constraints on the cosmic expansion history from GWTC-3
• Estimating the Contribution of Dynamical Ejecta in the Kilonova Associated with GW170817
• Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomy
• Enhancing Gravitational-Wave Science with Machine Learning
• Gravitational-wave parameter estimation with autoregressive neural network flows
• Neutron Star Equation of State in light of GW190814
• Learning Bayesian posteriors with neural networks for gravitational-wave inference
• Real-Time Detection of Gravitational Waves from Binary Neutron Stars using Artificial Neural Networks
• Gravitational Wave Denoising of Binary Black Hole Mergers with Deep Learning
• Reduced-order modeling with artificial neurons for gravitational-wave inference
• Deep-Learning continuous gravitational waves
• Gravitational wave signal recognition of O1 data by deep learning
• Enabling real-time multi-messenger astrophysics discoveries with deep learning
• Noise Reduction in Gravitational-wave Data via Deep Learning
• How effective is machine learning to detect long transient gravitational waves from neutron stars in a real search?
• Accelerated, Scalable and Reproducible AI-driven Gravitational Wave Detection
• Detection of gravitational-wave signals from binary neutron star mergers using machine learning
• Convergence of Artificial Intelligence and High Performance Computing on NSF-supported Cyberinfrastructure
• Applying deep neural networks to the detection and space parameter estimation of compact binary coalescence with a network of gravitational wave detectors
• Gravitational-wave surrogate models powered by artificial neural networks: The ANN-Sur for waveform generation
• Deep learning for clustering of continuous gravitational wave candidates
• Deep Learning Ensemble for Real-time Gravitational Wave Detection of Spinning Binary Black Hole Mergers
• Deep-Learning Continuous Gravitational Waves: Multiple detectors and realistic noise
• Deep learning for gravitational wave forecasting of neutron star mergers
• Training Strategies for Deep Learning Gravitational-Wave Searches
• Some Optimizations on Detecting Gravitational Wave Using Convolutional Neural Network
• Detection of Gravitational Waves Using Bayesian Neural Networks
• Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders
• Denoising Gravitational Waves using Deep Learning with Recurrent Denoising Autoencoders
• Fusing numerical relativity and deep learning to detect higher-order multipole waveforms from eccentric binary black hole mergers
• Early warning of coalescing neutron-star and neutron-star-black-hole binaries from nonstationary noise background using neural networks
• From One to Many: A Deep Learning Coincident Gravitational-Wave Search
• Deep Learning with Quantized Neural Networks for Gravitational Wave Forecasting of Eccentric Compact Binary Coalescence
• Interpretable AI forecasting for numerical relativity waveforms of quasi-circular, spinning, non-precessing binary black hole mergers
• Deep Learning Model on Gravitational Waveforms in Merging and Ringdown Phases of Binary Black Hole Coalescences
• Advances in Machine and Deep Learning for Modeling and Real-time Detection of Multi-Messenger Sources
• Nonlinear noise regression in gravitational-wave detectors with convolutional neural networks

Cited by in corpus (14)

• FAIR for AI: An interdisciplinary and international community building perspective
• MLGWSC-1: The first Machine Learning Gravitational-Wave Search Mock Data Challenge
• Data quality up to the third observing run of Advanced LIGO: Gravity Spy glitch classifications
• Applications of physics informed neural operators
• Deep Residual Networks for Gravitational Wave Detection
• Deep Learning Detection and Classification of Gravitational Waves from Neutron Star-Black Hole Mergers
• Gravity Spy: Lessons Learned and a Path Forward
• New Gravitational Wave Discoveries Enabled by Machine Learning
• Searches for Mass-Asymmetric Compact Binary Coalescence Events using Neural Networks in the LIGO/Virgo Third Observation Period
• AI and extreme scale computing to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non-precessing binary black hole mergers
• Novel Deep Learning Approach to Detecting Binary Black Hole Mergers
• Physics-inspired spatiotemporal-graph AI ensemble for the detection of higher order wave mode signals of spinning binary black hole mergers
• Conditional Autoencoder for Generating Binary Neutron Star Waveforms with Tidal and Precession Effects
• Sequence modeling of higher-order wave modes of binary black hole mergers