SourceNet: Interpretable Sim-to-Real Inference on Variable-Geometry Sensor Arrays for Earthquake Source Inversion

• URL: http://arxiv.org/abs/2601.06320v1
• Date: Fri, 09 Jan 2026 21:42:28 GMT
• Title: SourceNet: Interpretable Sim-to-Real Inference on Variable-Geometry Sensor Arrays for Earthquake Source Inversion
• Authors: Zhe Jia, Xiaotian Zhang, Junpeng Li,
• Abstract summary: SourceNet is a Transformer-based framework that treats the sensor array as a flexible set to model arbitrary geometries.<n>By pre-training on 100,000 synthetic events and fine-tuning on 2,000 real world events, SourceNet achieves state-of-the-art precision on held-out real data.
• Score: 7.936803576241796
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Inferring high-dimensional physical states from sparse, ad-hoc sensor arrays is a fundamental challenge across AI for Science, as they are complicated by irregular geometries and the profound Sim-to-Real gap in physical modeling. Taking earthquake source characterization as a representative challenge, we address limitations in conventional deep learning: CNNs demand fixed grids, while pooling-based architectures (e.g., DeepSets) struggle to capture the relational wave physics. Here, we propose SourceNet, a Transformer-based framework that treats the sensor array as a flexible set to model arbitrary geometries. To bridge the reality gap, we introduce Physics-Structured Domain Randomization (PSDR). Instead of forcing feature alignment, PSDR randomizes the governing physical dynamics by varying velocity structures, propagation effects, and sensor availability, to force the model to learn robust representations invariant to unmodeled environmental heterogeneity. By pre-training on 100,000 synthetic events and fine-tuning on ~2,000 real world events, SourceNet achieves state-of-the-art precision on held-out real data. This demonstrates exceptional data efficiency, and matches classical solvers while enabling real-time processing. Remarkably, interpretability analysis reveals that the model shows scientific-agent-like features: it autonomously discovers geometric information bottlenecks and learns an attention policy that prioritizes sparse sensor placements, effectively recovering principles of optimal experimental design from data alone.

Related papers

• Spatially-informed transformers: Injecting geostatistical covariance biases into self-attention for spatio-temporal forecasting [0.0]
  We propose a hybrid architecture that injects a geostatistic inductive bias directly into the decomposing self-attention mechanism via a learnable costatistics kernel.<n>We demonstrate the phenomenon of Deep Variography'', where the network successfully recovers the true spatial parameters of the underlying process end-to-end via backpropagation.
  arXiv  Detail & Related papers  (2025-12-19T15:32:24Z)
• XDen-1K: A Density Field Dataset of Real-World Objects [48.479432547763025]
  We introduce XDen-1K, the first dataset designed for real-world physical property estimation.<n>The core of this dataset consists of 1,000 real-world objects across 148 categories.<n>We introduce a novel optimization framework that recovers a high-fidelity volumetric density field of each object from its sparse X-ray views.
  arXiv  Detail & Related papers  (2025-12-11T14:15:42Z)
• Flexible Gravitational-Wave Parameter Estimation with Transformers [73.44614054040267]
  We introduce a flexible transformer-based architecture paired with a training strategy that enables adaptation to diverse analysis settings at inference time.<n>We demonstrate that a single flexible model -- called Dingo-T1 -- can analyze 48 gravitational-wave events from the third LIGO-Virgo-KAGRA Observing Run.
  arXiv  Detail & Related papers  (2025-12-02T17:49:08Z)
• Fusing CFD and measurement data using transfer learning [49.1574468325115]
  We introduce a non-linear method based on neural networks combining simulation and measurement data via transfer learning.<n>In a first step, the neural network is trained on simulation data to learn spatial features of the distributed quantities.<n>The second step involves transfer learning on the measurement data to correct for systematic errors between simulation and measurement by only re-training a small subset of the entire neural network model.
  arXiv  Detail & Related papers  (2025-07-28T07:21:46Z)
• Embedding Trust at Scale: Physics-Aware Neural Watermarking for Secure and Verifiable Data Pipelines [0.0]
  We present a robust neural watermarking framework for scientific data integrity.<n>Using a convolutional autoencoder, binary messages are invisibly embedded into structured data such as temperature, vorticity, and geopotential.<n>Our approach achieves $>$98% bit accuracy and visually indistinguishable reconstructions across ERA5 and Navier-Stokes datasets.
  arXiv  Detail & Related papers  (2025-05-22T21:14:45Z)
• Synth It Like KITTI: Synthetic Data Generation for Object Detection in Driving Scenarios [3.30184292168618]
  We propose a dataset generation pipeline based on the CARLA simulator for 3D object detection on LiDAR point clouds.<n>We are able to train an object detector on the synthetic data and demonstrate strong generalization capabilities to the KITTI dataset.
  arXiv  Detail & Related papers  (2025-02-20T22:27:42Z)
• Sparse identification of nonlinear dynamics and Koopman operators with Shallow Recurrent Decoder Networks [3.1484174280822845]
  We present a method to jointly solve the sensing and model identification problems with simple implementation, efficient, and robust performance.<n>SINDy-SHRED uses Gated Recurrent Units to model sparse sensor measurements along with a shallow network decoder to reconstruct the full-temporal field from the latent state space.<n>We conduct systematic experimental studies on PDE data such as turbulent flows, real-world sensor measurements for sea surface temperature, and direct video data.
  arXiv  Detail & Related papers  (2025-01-23T02:18:13Z)
• A Novel Diffusion Model for Pairwise Geoscience Data Generation with Unbalanced Training Dataset [8.453075713579631]
  We present UB-Diff'', a novel diffusion model for multi-modal paired scientific data generation.<n>One major innovation is a one-in-two-out encoder-decoder network structure, which can ensure pairwise data is obtained from a co-latent representation.<n> Experimental results on the OpenFWI dataset show that UB-Diff significantly outperforms existing techniques in terms of Fr'echet Inception Distance (FID) score and pairwise evaluation.
  arXiv  Detail & Related papers  (2025-01-01T19:49:38Z)
• GausSim: Foreseeing Reality by Gaussian Simulator for Elastic Objects [55.02281855589641]
  GausSim is a novel neural network-based simulator designed to capture the dynamic behaviors of real-world elastic objects represented through Gaussian kernels.<n>We leverage continuum mechanics and treat each kernel as a Center of Mass System (CMS) that represents continuous piece of matter.<n>In addition, GausSim incorporates explicit physics constraints, such as mass and momentum conservation, ensuring interpretable results and robust, physically plausible simulations.
  arXiv  Detail & Related papers  (2024-12-23T18:58:17Z)
• Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systems [49.11170948406405]
  We propose an unsupervised method to estimate the physical parameters of known, continuous governing equations from single videos.<n>We take the field closer to reality by recording Delfys75: our own real-world dataset of 75 videos for five different types of dynamical systems.
  arXiv  Detail & Related papers  (2024-10-02T09:44:54Z)
• PhyRecon: Physically Plausible Neural Scene Reconstruction [81.73129450090684]
  We introduce PHYRECON, the first approach to leverage both differentiable rendering and differentiable physics simulation to learn implicit surface representations. Central to this design is an efficient transformation between SDF-based implicit representations and explicit surface points. Our results also exhibit superior physical stability in physical simulators, with at least a 40% improvement across all datasets.
  arXiv  Detail & Related papers  (2024-04-25T15:06:58Z)
• DeepSimHO: Stable Pose Estimation for Hand-Object Interaction via Physics Simulation [81.11585774044848]
  We present DeepSimHO, a novel deep-learning pipeline that combines forward physics simulation and backward gradient approximation with a neural network. Our method noticeably improves the stability of the estimation and achieves superior efficiency over test-time optimization.
  arXiv  Detail & Related papers  (2023-10-11T05:34:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.