Metraipolis › archive › dc_4e75a0746983
Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation — arXiv:2608.10805
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10805
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:09:28.420Z
- SHA-256 of the archived text
28521641102b1b943fbfbc904d895e9a4f208f0646210053d7d34e27c2844d40- Stable URL of this page
https://metraipolis.com/documento/dc_4e75a0746983
How to cite
Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation — arXiv:2608.10805 — arXiv (en), https://arxiv.org/abs/2608.10805 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4e75a0746983 (Metraipolis archive dc_4e75a0746983, sha256 28521641102b1b943fbfbc904d895e9a4f208f0646210053d7d34e27c2844d40)
BibTeX
@misc{dc4e75a0746983,
title = {Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation — arXiv:2608.10805},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10805},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4e75a0746983; sha256 28521641102b1b943fbfbc904d895e9a4f208f0646210053d7d34e27c2844d40},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10805. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Wavelet convolution (WTConv) has emerged as an increasingly popular drop-in replacement for standard convolutions, expanding a network's receptive field exponentially with the number of decomposition levels while keeping the parameter count linear. However, its reference implementation is severely memory-bound due to excessive data movement through high-bandwidth memory (HBM). We develop an I/O model of WTConv to characterize this bottleneck and use it to guide three algebraic reformulations: (1) recomputing the inexpensive Haar analysis butterfly on chip, (2) collapsing the multi-level synthesis cascade into a single closed-form pass indexed by output-coordinate bits, and (3) folding learned per-channel scales into the convolution weights. Together, these reformulations enable an I/O-aware fused implementation that substantially reduces HBM traffic. We evaluate the WTConvNeXt configuration across decomposition levels and a broad range of tensor shapes. Despite performing comparable arithmetic, the reference WTConv is substantially slower than the depthwise convolution it replaces. Our reformulation reduces modeled HBM traffic by approximately $2.55\times$, yielding up to a $4.35\times$ training speedup over the reference while roughly halving peak memory usage. Thus, our reformulation preserves the benefits of WTConv while substantially reducing its execution time and memory footprint, removing the systems overhead that previously limited its practical efficiency. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.10805 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T05:57:28.406Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).