Introduction to Physics-Informed Neural Networks
Scientific machine learning
Physics-guided machine learning
Theory-guided neural networks
Neural-network approximation of physical fields
Universal approximation for differential equations
Incorporating physical laws into neural networks
Strong-form physics constraints
Weak-form physics constraints
Hard and soft physical constraints
Physics-informed loss functions
Residual-based learning
Automatic differentiation in PINNs
Computational graphs for physical derivatives
Neural-network trial solutions
Collocation-point methods
Mesh-free numerical modeling
Continuous-time PINNs
Discrete-time PINNs
Hybrid physics–data models
Ordinary differential equations with PINNs
Partial differential equations with PINNs
Integro-differential equations
Differential-algebraic equations
Stochastic differential equations
Fractional differential equations
Delay differential equations
Nonlinear differential equations
Coupled differential equations
Eigenvalue problems
Boundary-value problems
Initial-value problems
Initial-boundary-value problems
Forward problems
Inverse problems
Parameter-estimation problems
System-identification problems
Operator-learning problems
Multiscale differential equations
Singularly perturbed equations
Fully connected PINNs
Deep PINNs
Residual-network PINNs
Convolutional PINNs
Recurrent PINNs
Transformer-based PINNs
Graph neural-network PINNs
Fourier-feature PINNs
SIREN-based PINNs
Radial-basis-function PINNs
Bayesian PINNs
Probabilistic PINNs
Mixture-of-experts PINNs
Multi-network PINNs
Domain-decomposed PINNs
Multi-fidelity PINNs
Multi-output PINNs
Complex-valued PINNs
Symmetry-preserving PINNs
Equivariant PINNs
Governing-equation residual loss
Initial-condition loss
Boundary-condition loss
Data-matching loss
Conservation-law loss
Symmetry loss
Energy-conservation loss
Momentum-conservation loss
Mass-conservation loss
Charge-conservation loss
Entropy-condition loss
Interface-condition loss
Integral-constraint loss
Regularization terms
Multi-objective loss functions
Adaptive loss weighting
Dynamic loss balancing
Gradient-based loss balancing
Augmented Lagrangian constraints
Penalty-method constraints
Dirichlet boundary conditions
Neumann boundary conditions
Robin boundary conditions
Periodic boundary conditions
Mixed boundary conditions
Absorbing boundary conditions
Radiation boundary conditions
Interface boundary conditions
Moving-boundary conditions
Free-boundary problems
Exact enforcement of boundary conditions
Approximate enforcement of boundary conditions
Boundary-embedded neural architectures
Initial-condition embedding
Irregular-domain boundary treatment
Gradient-descent training
Adam optimization for PINNs
L-BFGS optimization
Second-order optimization
Natural-gradient methods
Hybrid Adam–L-BFGS training
Learning-rate scheduling
Transfer learning for PINNs
Curriculum learning
Causal training
Sequential time-domain training
Time marching with PINNs
Adaptive activation functions
Weight initialization
Gradient clipping
Gradient normalization
Vanishing-gradient problems
Exploding-gradient problems
Spectral bias in PINNs
Training instability
Uniform collocation sampling
Random collocation sampling
Latin hypercube sampling
Sobol-sequence sampling
Quasi-Monte Carlo sampling
Adaptive collocation sampling
Residual-based adaptive refinement
Importance sampling
Boundary-focused sampling
Interface-point sampling
Time-adaptive sampling
Error-driven sampling
Active-learning PINNs
Dynamic resampling
Collocation-point density optimization
Variational PINNs
Weak-form PINNs
Conservative PINNs
Extended PINNs
Parallel PINNs
Distributed PINNs
Finite-basis PINNs
Deep Ritz methods
Deep Galerkin methods
Physics-informed DeepONets
Fourier neural operators with physics constraints
Physics-informed neural operators
Gradient-enhanced PINNs
Conservative physics-informed neural networks
Domain-decomposition PINNs
XPINNs
hp-VPINNs
cPINNs
gPINNs
Bayesian physics-informed neural networks
Newtonian motion
Harmonic oscillator modeling
Damped oscillators
Driven oscillators
Coupled oscillators
Pendulum dynamics
Double-pendulum dynamics
Rigid-body motion
Rotational dynamics
Lagrangian mechanics
Hamiltonian mechanics
Hamiltonian neural networks
Symplectic PINNs
Chaotic dynamical systems
Three-body problems
Orbital dynamics
Celestial mechanics
Contact mechanics
Friction modeling
Impact and collision dynamics
Navier–Stokes equations
Euler equations
Incompressible fluid flow
Compressible fluid flow
Laminar flow
Turbulent flow
Boundary-layer flow
Channel flow
Pipe flow
Couette flow
Poiseuille flow
Lid-driven cavity flow
Flow around cylinders
Aerodynamic flow
Supersonic flow
Hypersonic flow
Multiphase flow
Porous-media flow
Non-Newtonian fluids
Magnetohydrodynamic flow
Free-surface flow
Fluid–structure interaction
Shock-wave modeling
Rarefied gas dynamics
Turbulence closure with PINNs
Heat equation
Steady-state heat conduction
Transient heat conduction
Convection–diffusion equations
Radiative heat transfer
Conjugate heat transfer
Phase-change problems
Stefan problems
Thermal diffusion
Nonlinear heat transfer
Thermoelasticity
Thermodynamic state reconstruction
Entropy-constrained neural networks
Energy-balance modeling
Thermal parameter estimation
Maxwell’s equations with PINNs
Electrostatic field modeling
Magnetostatic field modeling
Electromagnetic-wave propagation
Waveguide modeling
Antenna-field reconstruction
Electromagnetic inverse scattering
Maxwell–Bloch equations
Plasma-wave modeling
Magnetohydrodynamics
Vlasov equations
Vlasov–Poisson equations
Particle-in-cell surrogate modeling
Magnetic reconnection
Plasma transport
Charged-particle dynamics
Space-plasma modeling
Ionospheric plasma reconstruction
Fusion-plasma modeling
Tokamak equilibrium reconstruction
Schrödinger equation
Time-dependent Schrödinger equation
Time-independent Schrödinger equation
Quantum harmonic oscillator
Quantum tunneling
Quantum scattering
Quantum bound states
Quantum eigenvalue problems
Many-body quantum systems
Density-functional equations
Gross–Pitaevskii equation
Bose–Einstein condensates
Nonlinear Schrödinger equation
Quantum control
Quantum-state reconstruction
Quantum Hamiltonian identification
Open quantum systems
Lindblad master equations
Quantum field reconstruction
Physics-informed quantum neural networks
Special-relativistic dynamics
General-relativistic field equations
Einstein field equations
Geodesic equations
Schwarzschild spacetime
Kerr spacetime
Gravitational-wave equations
Numerical-relativity surrogates
Cosmological evolution equations
Friedmann equations
Relativistic hydrodynamics
Black-hole parameter estimation
Gravitational lensing inversion
Spacetime metric reconstruction
Modified-gravity models
Stellar-structure equations
Stellar-evolution modeling
Hydrostatic equilibrium in stars
Radiative-transfer equations
Accretion-disk modeling
Supernova hydrodynamics
Neutron-star modeling
Compact-object dynamics
Cosmic-ray transport
Neutrino transport
Dark-matter distribution reconstruction
Dark-energy models
Large-scale structure formation
Cosmological parameter estimation
Galaxy-dynamics modeling
Interstellar-medium modeling
Solar-physics applications
Helioseismic inversion
Space-weather forecasting
Exoplanet atmospheric modeling
Nuclear reaction modeling
Nuclear-decay equations
Neutron-transport equations
Reactor-physics modeling
Nuclear diffusion equations
Nuclear cross-section estimation
Particle-transport equations
Boltzmann transport equations
Relativistic collision dynamics
Detector-response reconstruction
Particle-track reconstruction
Calorimeter-response modeling
Neutrino oscillation equations
Neutrino tomography
High-energy air-shower modeling
Cosmic-ray propagation
Accelerator beam dynamics
Plasma-wakefield acceleration
Lattice-field equations
Quantum chromodynamics surrogate models
Elasticity equations
Plasticity modeling
Solid mechanics
Continuum mechanics
Stress–strain reconstruction
Fracture mechanics
Crack propagation
Damage mechanics
Phase-field modeling
Crystal plasticity
Dislocation dynamics
Microstructure evolution
Diffusion in solids
Semiconductor-device equations
Poisson–Schrödinger systems
Superconductivity equations
Ginzburg–Landau equations
Spin dynamics
Micromagnetics
Materials-property prediction
Seismic-wave propagation
Seismic inversion
Earthquake-source reconstruction
Elastic-wave equations
Acoustic-wave equations
Full-waveform inversion
Geothermal modeling
Groundwater flow
Soil-moisture transport
Subsurface-flow reconstruction
Reservoir simulation
Petroleum-system modeling
Geomagnetic-field modeling
Mantle-convection modeling
Plate-tectonic dynamics
Volcanic-flow modeling
Tsunami-wave modeling
Ocean-current modeling
Atmospheric-flow modeling
Climate-system modeling
Wave equation
Helmholtz equation
Acoustic-wave propagation
Sound-field reconstruction
Room acoustics
Ultrasound propagation
Optical-wave propagation
Nonlinear optics
Fiber-optic modeling
Photonic-crystal modeling
Laser dynamics
Inverse optical scattering
Diffraction modeling
Interference modeling
Metamaterial modeling
Seismic and acoustic tomography
Wave-source localization
Wavefront reconstruction
Electromagnetic tomography
Photoacoustic imaging
Unknown-parameter identification
Unknown-source identification
Boundary-condition inference
Initial-condition inference
Hidden-state reconstruction
Coefficient inversion
Constitutive-law discovery
PDE discovery from data
Governing-equation identification
Sparse-data reconstruction
Noisy-data inversion
Sensor-data assimilation
Field reconstruction
Tomographic reconstruction
Source localization
Material-property inversion
Model calibration
State estimation
Online system identification
Digital-twin calibration
Uncertainty quantification in PINNs
Aleatoric uncertainty
Epistemic uncertainty
Bayesian inference
Monte Carlo PINNs
Ensemble PINNs
Probabilistic parameter estimation
Confidence intervals for physical fields
Robust PINNs
Noise-resistant PINNs
Out-of-distribution detection
Sensitivity analysis
Error propagation
Reliability analysis
Risk-aware physics-informed learning
Analytical-solution comparison
Numerical-solution comparison
Finite-difference benchmarking
Finite-element benchmarking
Finite-volume benchmarking
Spectral-method benchmarking
Residual-error analysis
Relative (L_2)-error analysis
Conservation-error analysis
Boundary-condition error
Generalization across parameters
Extrapolation in time
Extrapolation in space
Convergence analysis
Stability analysis
Computational-cost analysis
Accuracy–efficiency trade-offs
Ablation studies
Reproducibility of PINN models
Benchmark PDE datasets
PINNs with TensorFlow
PINNs with PyTorch
PINNs with JAX
PINNs with DeepXDE
PINNs with NVIDIA Modulus
PINNs with SciANN
GPU-accelerated PINNs
Multi-GPU PINNs
Distributed computing
Parallel automatic differentiation
Mixed-precision training
Memory-efficient PINNs
High-performance computing for PINNs
Differentiable programming
Symbolic–neural computation
PINN software architecture
Scientific-data preprocessing
Visualization of PINN solutions
Reproducible PINN workflows
Open-source PINN benchmarks
Physics-informed foundation models
Physics-informed generative models
Physics-informed diffusion models
Physics-informed transformers
Physics-informed graph networks
Physics-informed reinforcement learning
Physics-informed operator learning
Neural PDE solvers
Equation-free physics learning
Automated scientific discovery
Symbolic regression with PINNs
Differentiable simulators
Physics-informed digital twins
Real-time physical forecasting
Experimental control using PINNs
Sensor-placement optimization
Multi-physics PINNs
Multi-scale PINNs
Multi-domain PINNs
Foundation models for scientific simulation
Spectral bias
Gradient pathologies
Stiff differential equations
High-dimensional PDEs
Long-time integration
Chaotic-system prediction
Shock and discontinuity representation
Complex geometries
Irregular boundaries
Sparse observational data
Noisy experimental data
Conflicting physical constraints
Loss-function imbalance
Computational scalability
Accuracy guarantees
Convergence guarantees
Interpretability
Physical consistency
Generalization beyond the training domain
Comparison with conventional numerical solvers
Experimental-data reconstruction
Sensor fusion
Reduced-order modeling
Surrogate modeling
Digital twins of physical systems
Real-time monitoring
Fault detection
Predictive maintenance
Structural-health monitoring
Process control
Robotics dynamics
Aerospace-system modeling
Energy-system optimization
Battery modeling
Fuel-cell modeling
Renewable-energy forecasting
Medical-physics modeling
Biophysical-system modeling
Laboratory experiment design
Autonomous scientific experimentation
These topics cover the mathematical foundations, architectures, numerical techniques, physical applications, software implementation, inverse problems, uncertainty analysis, and current research challenges associated with Physics-Informed Neural Networks.