Expert: 'If you find a deep learning method that outperforms a linear model, I buy you a drink'
danilobzdok · x · 2026-08-21
Martin Lindquist stated at the IVADO workshop that he would buy a drink for anyone who could find a deep learning method that outperforms a linear model. This echoes the author's 2020 benchmarking study in Nature Communications, which found that linear models often outperform deep learning on datasets like UKBiobank brain images due to sample size constraints.
Related event: fMRI Expert Offers Drinks If Deep Learning Beats Linear Models(2 posts)→
More from Research
- Stanford prof Kundaje: a smoothly running auto-orchestrator is a big unlock — anshulkundaje · 2026-08-21
- Researcher publishes 8 months of primary-source records on frontier model behavior — rayanpal_ · 2026-08-21
- Ex-Stability researcher seeks funding to test AI-assisted attacks on a wall of everyday hardware — hugobowne · 2026-08-21
- NeurIPS 2026 Queer in AI x {Dis}ability in AI workshop: submissions due Sep 3 — mundt_martin · 2026-08-21
- Hydra-0: a generalist world model that represents robot actions as action flow — mangahomanga · 2026-08-21
- Does Claude perform better in 'claudish'? Researchers call for empirical measures — voooooogel · 2026-08-21