In Science Labs, 200 Data Points Often Need a Gaussian Process, Not Deep Learning
bravo_abad · x · 2026-09-07
The author revisits the classic nesting—AI contains ML, ML contains deep learning—and argues the interesting layer in scientific labs is often the middle ring, not deep learning.
- Experiments with 200 data points frequently call for Gaussian processes, random forests, or regularized regression instead of deep networks.
- Error bars matter most: they determine what you measure next.
- Takeaway: in AI for Science, "classical" isn't outdated—it's often the right fit.
More from Research
- scikit-learn 1.9 ships metric_at_thresholds to simplify optimal decision threshold search — GaelVaroquaux · 2026-09-08
- Astra agent inside Codex picks the same cancer sequencing variants a researcher would choose — iskander · 2026-09-08
- Enterprise agent evals need world-first design, not task-first, argues Shahules Anwar — Shahules786 · 2026-09-08
- Ineffable Labs adds six co-founders alongside ex-DeepMind's David Silver — giffmana · 2026-09-08
- Training a 9B model with GRPO to build low-poly Blender rooms: lessons learned — TheMoonMidas · 2026-09-08
- 2026 PNPL competition targets non-invasive speech decoding with MEG — pnpl · 2026-09-08