Solving differential equations with neural networks: Applications to the calculation of cosmological phase transitions
arXiv:1902.05563 · doi:10.1103/PhysRevD.100.016002
Abstract
Starting from the observation that artificial neural networks are uniquely suited to solving optimisation problems, and most physics problems can be cast as an optimisation task, we introduce a novel way of finding a numerical solution to wide classes of differential equations. We find our approach to be very flexible and stable without relying on trial solutions, and applicable to ordinary, partial and coupled differential equations. We apply our method to the calculation of tunnelling profiles for cosmological phase transitions, which is a problem of relevance for baryogenesis and stochastic gravitational wave spectra. Comparing our solutions with publicly available codes which use numerical methods optimised for the calculation of tunnelling profiles, we find our approach to provide at least as accurate results as these dedicated differential equation solvers, and for some parameter choices even more accurate and reliable solutions. In particular, we compare the neural network approach with two publicly available profile solvers, \texttt{CosmoTransitions} and \texttt{BubbleProfiler}, and give explicit examples where the neural network approach finds the correct solution while dedicated solvers do not. We point out that this approach of using artificial neural networks to solve equations is viable for any problem that can be cast into the form , and is thus applicable to various other problems in perturbative and non-perturbative quantum field theory.
arXiv comment12 pages, 9 figures. V2: Typo fixes, title changed, agrees with version published in Phys. Rev. D
References in corpus (44)
- The Minimal Model of Nonbaryonic Dark Matter: A Singlet Scalar
- Science with the space-based interferometer eLISA. II: Gravitational waves from cosmological phase transitions
- Electroweak baryogenesis
- Electroweak Baryon Number Non-Conservation in the Early Universe and in High Energy Collisions
- CosmoTransitions: Computing Cosmological Phase Transition Temperatures and Bubble Profiles with Multiple Fields
- Thermally Generated Gauge Singlet Scalars as Self-Interacting Dark Matter
- Singlet Higgs Phenomenology and the Electroweak Phase Transition
- Gravitational Waves from Phase Transitions at the Electroweak Scale and Beyond
- Electroweak Baryogenesis and Standard Model CP Violation
- Strong Electroweak Phase Transitions in the Standard Model with a Singlet
- Sterile neutrinos, dark matter, and the pulsar velocities in models with a Higgs singlet
- Classification without labels: Learning from mixed samples in high energy physics
- Deep learning in color: towards automated quark/gluon jet discrimination
- Jet-Images -- Deep Learning Edition
- Gravitational Waves from Electroweak Phase Transitions
- Deep-learning Top Taggers or The End of QCD?
- A determination of parton distributions with faithful uncertainty estimation
- Anomaly Detection for Resonant New Physics with Machine Learning
- Electroweak baryogenesis and dark matter from a singlet Higgs
- Learning New Physics from a Machine
- Deep-learned Top Tagging with a Lorentz Layer
- Constraining Effective Field Theories with Machine Learning
- Electroweak Baryogenesis: Concrete in a SUSY Model with a Gauge Singlet
- A unified explanation for dark matter and electroweak baryogenesis with direct detection and gravitational wave signatures
- What is the Machine Learning?
- Comments on the Electroweak Phase Transition
- Parton Shower Uncertainties in Jet Substructure Analyses with Deep Neural Networks
- Electroweak Phase Transition and Baryogenesis in the nMSSM
- Bubbles in the supersymmetric standard model
- Search for the Standard Model Higgs boson produced in association with top quarks and decaying into a pair in collisions at = 13 TeV with the ATLAS detector
- Efficient numerical solution to vacuum decay with many fields
- Was the Electroweak Phase Transition Preceded by a Color-Broken Phase?
- Machine Learning Uncertainties with Adversarial Neural Networks
- Numerical Approach to Multi Dimensional Phase Transitions
- Search for the Higgs boson decaying to two muons in proton-proton collisions at 13 TeV
- A Fresh Look at the Calculation of Tunneling Actions
- Bubble Wall Profiles With More Than One Scalar Field: A Numerical Approach
- A Fresh Look at the Calculation of Tunneling Actions in Multi-Field Potentials
- Semi-analytic techniques for calculating bubble wall profiles
- Multifield Polygonal Bounces
- Constrained potential method for false vacuum decays
- Search for supersymmetry in events with opposite-sign dileptons and missing transverse energy using an artificial neural network
- Signs of heavy Higgs bosons at CLIC: An road to the Electroweak Phase Transition
- Double-charming Higgs identification using machine-learning assisted jet shapes
Cited by in corpus (25)
- Collider and Gravitational Wave Complementarity in Exploring the Singlet Extension of the Standard Model
- Deep Learning for Feynman's Path Integral in Strong-Field Time-Dependent Dynamics
- Quantum Machine Learning for Particle Physics using a Variational Quantum Classifier
- FindBounce: package for multi-field bounce actions
- Effective field theory approach to thermal bubble nucleation
- Cosmological bubble friction in local equilibrium
- Computing the gauge-invariant bubble nucleation rate in finite temperature effective field theory
- BubbleDet: A Python package to compute functional determinants for bubble nucleation
- Quantum Computing for Quantum Tunnelling
- Hyperbolic compactification of M-theory and de Sitter quantum gravity
- Quasinormal modes of dS and AdS Black Holes: Feedforward neural network method
- Quantum Optimisation of Complex Systems with a Quantum Annealer
- Simple Gradient Flow Equation for the Bounce Solution
- Using Machine Learning to disentangle LHC signatures of Dark Matter candidates
- Cosmology-informed neural networks to solve the background dynamics of the Universe
- Hypersolvers: Toward Fast Continuous-Depth Models
- Elvet -- a neural network-based differential equation and variational problem solver
- Machine Learning Etudes in Conformal Field Theories
- Optimisation based algorithm for finding the action of cosmological phase transitions
- The Emergence of Electroweak Skyrmions through Higgs Bosons
- Efficient time stepping for numerical integration using reinforcement learning
- meta-stable states in PNJL model
- Electroweak Skyrmions in the HEFT
- Quartic Gradient Flow
- Unsupervised Learning of Solutions to Differential Equations with Generative Adversarial Networks