Loss-Conditional PINNs for Parametric PDE Families
About
Physics-informed neural networks (PINNs) approximate solutions of ODEs and PDEs by minimising a weighted combination of residual, boundary, initial, and data losses. Their performance is often dominated by the choice of loss weights: a poor weighting can drive training to a degenerate solution in which one physical constraint is satisfied while another is ignored. Existing methods select or adapt a single good set of weights. We take a different view: instead of tuning one weight vector, we explore the entire weight space during training. We introduce LC-PINN, which adapts the loss-conditional training of Dosovitskiy and Djolonga (2020) to the PDE-residual setting: the conditioning vector (either the loss weights or a scalar physical coefficient) is treated as a network input and sampled from a simple prior at every optimisation step. This turns PINN training into learning a continuous family of solutions indexed by that vector, with no solver-generated paired data. LC-PINN thus lies between classical PINNs and operator learning: it stays fully physics-informed but amortises training over a parametric family. Our contribution is not the loss-conditional construction itself, but its extension to PINNs, the unification of the loss-weight and parametric-coefficient regimes under one architecture (concatenation for loss weights, FiLM for coefficients), and a fixed-quadrature L-BFGS finishing protocol that makes the parametric-coefficient regime trainable. We give a lambda-invariance result for the conditional optimum and study LC-PINN on parametric Helmholtz, Schrodinger, viscous Burgers, and Buckley-Leverett equations. A single LC-PINN matches or improves retrained per-weight PINN baselines while parameterising the full family in one model, at a total cost that amortises favourably against per-instance retraining.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Parametric PDE Solving | Viscous Burgers equation initial condition u(0, x) = -sin(πx) v=0.01/π (test) | Rel L2 Error (Test)3.47 | 5 | |
| Solving Parametric PDEs | 1D Helmholtz (k ∈ {1.00, 3.25, 5.50, 7.75, 10.00}) | Relative L2 Error9.39 | 4 | |
| Parametric PDE Solving | Viscous-regularized Buckley–Leverett (ε = 10−2) (test snapshots) | Relative L2 Error1.03 | 3 | |
| Parametric PDE Solving | 1D Schrödinger equation with parametric harmonic well (alpha in {0.5, 5.0, 10.0}) (test) | Rel L2 Error (Mean)1.99 | 3 | |
| Solving 3D Helmholtz PDE | 3D Helmholtz on the unit cube k ∈ {1, 3, 5} | Relative L2 Error1.67 | 3 | |
| Solving Parametric Partial Differential Equations | 2D Helmholtz on the unit square k ∈ {1.00, 5.50, 10.00} | Relative L2 Error (mean)2.36 | 3 |