Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper investigates cross-Reynolds generalization in neural PDE solvers, specifically focusing on the 2D Navier-Stokes equation. They find that representation geometry, rather than just forward model accuracy, is a key factor, as evidenced by the performance of a ConvAE-Relay method that matches states in a source-trained latent space. Through ablations and oracle experiments, the authors demonstrate that matching quality and autoregressive drift are critical bottlenecks, and that local, multi-scale representations facilitate cross-Reynolds transfer.
Surprisingly, simply retrieving dynamics from a source-regime database via latent space matching outperforms a trained Fourier Neural Operator in cross-Reynolds generalization for 2D Navier-Stokes, highlighting the importance of representation geometry.
Cross-Reynolds generalisation in neural PDE solvers remains poorly characterised. On the canonical forced 2D Navier-Stokes benchmark, a trained Fourier Neural Operator reaches 46.68% relative L2 error under a 10x Reynolds-number shift, yet zero-forward-model retrieval baselines already improve to 41-42%. This suggests representation geometry as a major organising variable among the tested methods. We test this hypothesis through ConvAE-Relay, which matches states in a source-trained convolutional autoencoder latent space and borrows dynamics from a source-regime database, achieving 38.34+/-0.07% using only a source-regime database and no target-regime fitting, labels, or database entries. A 2x2 ablation isolates matching quality as dominant over the update rule. Oracle experiments confirm that source-regime dynamics directions remain transferable (cosine similarity ~0.84) when matching stays on-manifold; autoregressive drift is the primary bottleneck (~12 percentage points). From the learned-prediction side, a U-Net with multi-scale skip connections achieves 34.72+/-0.60%, consistent with the retrieval-side finding that local, multi-scale representations organise cross-Reynolds transfer among tested methods. All claims are scoped to this benchmark.