Meta learned multi exit method improves seismic data interpolation efficiency
MetaPOCS: Meta-Learned Multi-Exit Unfolding for Budget-Aware Seismic Data Interpolation
Machine Learning
Summary
Seismic data interpolation helps fill in missing information in underground maps, but doing so can be very slow or require lots of computer power. The authors present MetaPOCS, a new approach that learns to produce useful partial results at different stages, allowing for faster or more precise outputs depending on available computing budget. This method trains the system to improve intermediate guesses, so it can stop early when needed without losing quality. Their approach also balances accuracy and speed better than previous fixed-step methods.
What this means in practice
- •For geophysical data engineers: Run seismic interpolation models that adapt to varied computational budgets by stopping early or running fully, improving processing time and output quality in exploration workflows.
- •For medical imaging developers: Implement multi-exit unfolding techniques to speed up image reconstruction tasks with adjustable accuracy, supporting real-time diagnostics under computational constraints.
Authors
Baisong Jiang, Chunxia Zhang, Xiaoli Wei, Anxiang Di, Nanqi Zhao, Deng Xiong
Abstract
Seismic data interpolation must reconcile reconstruction fidelity with the computational budgets of acquisition quality control and subsequent imaging. Fixed-depth unfolding networks are optimized for a single operating point, whereas their intermediate estimates may be insufficiently trained for early deployment. We present MetaPOCS, a meta-learned multi-exit unfolding framework for budget-aware seismic interpolation. Building on a spatial--spectral POCS reconstruction backbone, MetaPOCS supervises intermediate stages and preserves the acquired traces at every exit. A Meta-GF-based optimization procedure constructs exit-oriented parameter displacements from a common initialization and learns their fusion weights through an equal-exit outer objective. The weights vary across parameter tensors and respect the nested support of the reconstruction stages. We derive the resulting fusion hypergradient to characterize how the outer objective reallocates competing updates, and distinguish this mechanism from scalar loss reweighting. At inference, one checkpoint provides a set of reconstruction budgets through a latency-calibrated exit policy. The evaluation framework couples matched-cost interpolation comparisons with finite-update diagnostics and a fixed reverse-time-migration operator, connecting multi-exit optimization to waveform and imaging fidelity. MetaPOCS thus provides a unified formulation for studying the reconstruction--computation trade-off of seismic unfolding across deployment budgets.