analiza aiqbench
This study demonstrates that large-scale synthetic pretraining using open-source ray tracing can significantly enhance indoor radio-map prediction models, reducing error by 15% even with simulator mismatch. By establishing predictable data-scaling trends across ResNet and Swin-T architectures, the work provides a practical pathway to overcome the scarcity of real-world measurement data in wireless network planning.
Scaling Cheap Synthetic Pretraining Improves Indoor Radio-Map Prediction
arXiv:2610.19015v1 Announce Type: new Abstract: Real measurements are too limited to train an accurate indoor pathloss predictor that works across a wide range of buildings. Large-scale synthetic pretraining offers a way forward, but real building layouts are also scarce, and accurate ray-tracing simulators are slow, expensive, and often proprietary. We address both problems by introducing a generator that automatically creates diverse indoor layouts and a fast, open-source approximate ray tracer that computes radio maps for them. Using this pipeline, we generate 128M synthetic samples, pretrain ResNet-101 and Swin-T encoder-decoders, and fine-tune them on high-quality benchmark data. Despite simulator mismatch, our ResNet-101 model reduces error by 15% on average relative to the best known methods across five indoor radio-map prediction tasks and improves transfer on a small real-measurement dataset. Both architectures show predictable data-scaling trends: power laws fitted on 2M-16M samples predict the mean normalized downstream RMSE at 32M-128M with 3.33% mean absolute percentage error for ResNet-101 and 7.10% for Swin-T. These results show that pretraining on large amounts of cheap simulated data can improve indoor radio-map prediction, even when real measurements and high-quality simulations are limited.
Abstract:Real measurements are too limited to train an accurate indoor pathloss predictor that works across a wide range of buildings. Large-scale synthetic pretraining offers a way forward, but real building layouts are also scarce, and accurate ray-tracing simulators are slow, expensive, and often proprietary. We address both problems by introducing a generator that automatically creates diverse indoor layouts and a fast, open-source approximate ray tracer that computes radio maps for them. Using this pipeline, we generate 128M synthetic samples, pretrain ResNet-101 and Swin-T encoder-decoders, and fine-tune them on high-quality benchmark data. Despite simulator mismatch, our ResNet-101 model reduces error by 15% on average relative to the best known methods across five indoor radio-map prediction tasks and improves transfer on a small real-measurement dataset. Both architectures show predictable data-scaling trends: power laws fitted on 2M-16M samples predict the mean normalized downstream RMSE at 32M-128M with 3.33% mean absolute percentage error for ResNet-101 and 7.10% for Swin-T. These results show that pretraining on large amounts of cheap simulated data can improve indoor radio-map prediction, even when real measurements and high-quality simulations are limited.
Accepted at NeurIPS 2026. This research was in part sponsored by the HESC MESCS RA grant No. 22rl-052 (DISTAL), and by the NATO Science for Peace and Security Programme under grant ID G9270 (OASIS)
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.19015 [cs.LG]
(or arXiv:2610.19015v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.19015 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 236 cuvinte
Subiecte conexe
Research Note: Filtering Subversion-Relevant Information From Pretraining Data Is Feasible
LessWrong · 5 oct. 2026
PIRAMID: Progress and Plans
LessWrong · 27 iul. 2026
Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Import AI · 28 sept. 2026
Xiaomi-TabLDM: A Tabular Foundation Model Technical Report
arXiv · 5 sept. 2026