ANTiC: Adaptive Neural Temporal In Situ Compressor

Sandeep Suresh Cranganore, Andrei Bodnar, Gianluca Galletti, Fabian Paischer, Johannes Brandstetter
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:21650-21680, 2026.

Abstract

The persistent storage requirements for high-resolution, spatiotemporally evolving fields governed by large-scale and high-dimensional partial differential equations (PDEs) have reached the petabyte-to-exabyte scale. Transient simulations modeling Navier-Stokes equations, magnetohydrodynamics, plasma physics, or binary black hole mergers generate data volumes that are prohibitive for modern high-performance computing (HPC) infrastructures. To address this bottleneck, we introduce ANTIC (Adaptive Neural Temporal in situ Compressor), an end-to-end in situ compression pipeline. ANTIC consists of an adaptive temporal selector tailored to high-dimensional physics that identifies and filters informative snapshots at simulation time, combined with a spatial neural compression module based on continual fine-tuning that learns residual updates between adjacent snapshots using neural fields. By operating in a single streaming pass, ANTIC enables a combined compression of temporal and spatial components and effectively alleviates the need for explicit on-disk storage of entire time-evolved trajectories. Experimental results demonstrate that ANTIC achieves storage reductions of approximately $\sim 400\times$ for 2D Kolmogorov flow simulations and $\sim 7000\times$ for large-scale physics simulations such as binary black hole mergers.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-cranganore26a, title = {{ANT}i{C}: Adaptive Neural Temporal In Situ Compressor}, author = {Cranganore, Sandeep Suresh and Bodnar, Andrei and Galletti, Gianluca and Paischer, Fabian and Brandstetter, Johannes}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {21650--21680}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/cranganore26a/cranganore26a.pdf}, url = {https://proceedings.mlr.press/v306/cranganore26a.html}, abstract = {The persistent storage requirements for high-resolution, spatiotemporally evolving fields governed by large-scale and high-dimensional partial differential equations (PDEs) have reached the petabyte-to-exabyte scale. Transient simulations modeling Navier-Stokes equations, magnetohydrodynamics, plasma physics, or binary black hole mergers generate data volumes that are prohibitive for modern high-performance computing (HPC) infrastructures. To address this bottleneck, we introduce ANTIC (Adaptive Neural Temporal in situ Compressor), an end-to-end in situ compression pipeline. ANTIC consists of an adaptive temporal selector tailored to high-dimensional physics that identifies and filters informative snapshots at simulation time, combined with a spatial neural compression module based on continual fine-tuning that learns residual updates between adjacent snapshots using neural fields. By operating in a single streaming pass, ANTIC enables a combined compression of temporal and spatial components and effectively alleviates the need for explicit on-disk storage of entire time-evolved trajectories. Experimental results demonstrate that ANTIC achieves storage reductions of approximately $\sim 400\times$ for 2D Kolmogorov flow simulations and $\sim 7000\times$ for large-scale physics simulations such as binary black hole mergers.} }
Endnote
%0 Conference Paper %T ANTiC: Adaptive Neural Temporal In Situ Compressor %A Sandeep Suresh Cranganore %A Andrei Bodnar %A Gianluca Galletti %A Fabian Paischer %A Johannes Brandstetter %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-cranganore26a %I PMLR %P 21650--21680 %U https://proceedings.mlr.press/v306/cranganore26a.html %V 306 %X The persistent storage requirements for high-resolution, spatiotemporally evolving fields governed by large-scale and high-dimensional partial differential equations (PDEs) have reached the petabyte-to-exabyte scale. Transient simulations modeling Navier-Stokes equations, magnetohydrodynamics, plasma physics, or binary black hole mergers generate data volumes that are prohibitive for modern high-performance computing (HPC) infrastructures. To address this bottleneck, we introduce ANTIC (Adaptive Neural Temporal in situ Compressor), an end-to-end in situ compression pipeline. ANTIC consists of an adaptive temporal selector tailored to high-dimensional physics that identifies and filters informative snapshots at simulation time, combined with a spatial neural compression module based on continual fine-tuning that learns residual updates between adjacent snapshots using neural fields. By operating in a single streaming pass, ANTIC enables a combined compression of temporal and spatial components and effectively alleviates the need for explicit on-disk storage of entire time-evolved trajectories. Experimental results demonstrate that ANTIC achieves storage reductions of approximately $\sim 400\times$ for 2D Kolmogorov flow simulations and $\sim 7000\times$ for large-scale physics simulations such as binary black hole mergers.
APA
Cranganore, S.S., Bodnar, A., Galletti, G., Paischer, F. & Brandstetter, J.. (2026). ANTiC: Adaptive Neural Temporal In Situ Compressor. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:21650-21680 Available from https://proceedings.mlr.press/v306/cranganore26a.html.

Related Material