{"ID":2844736,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.06157","arxiv_id":"2511.06157","title":"Models Got Talent: Identifying High Performing Wearable Human Activity Recognition Models Without Training","abstract":"A promising alternative to the computationally expensive Neural Architecture Search (NAS) involves the development of Zero Cost Proxies (ZCPs), which correlate well with trained performance, but can be computed through a single forward/backward pass on a randomly sampled batch of data. In this paper, we investigate the effectiveness of ZCPs for HAR on six benchmark datasets, and demonstrate that they discover network architectures that obtain within 5% of performance attained by full-scale training involving 1500 randomly sampled architectures. This results in substantial computational savings as high-performing architectures can be discovered with minimal training. Our experiments not only introduce ZCPs to sensor-based HAR, but also demonstrate that they are robust to data noise, further showcasing their suitability for practical scenarios.","short_abstract":"A promising alternative to the computationally expensive Neural Architecture Search (NAS) involves the development of Zero Cost Proxies (ZCPs), which correlate well with trained performance, but can be computed through a single forward/backward pass on a randomly sampled batch of data. In this paper, we investigate the...","url_abs":"https://arxiv.org/abs/2511.06157","url_pdf":"https://arxiv.org/pdf/2511.06157v2","authors":"[\"Richard Goldman\",\"Varun Komperla\",\"Thomas Ploetz\",\"Harish Haresamudram\"]","published":"2025-11-08T22:38:14Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[]","has_code":false}