MICCAI Researcher: Recent NeurIPS Spectral Learning Papers Ignore the Strasbourg Gamma-Calculus Tradition
PTenigma · x · 2026-09-28
A researcher posting ahead of MICCAI in Strasbourg notes that the city's math tradition — Meyer of Doob-Meyer decomposition fame, plus Bakry and Emery, who developed Gamma and Gamma₂ calculus there over 50+ years — essentially discovered multiple descent in deep learning, yet recent NeurIPS papers on the spectral theory of learning use these tools without citing or knowing the Strasbourg school.
He invites attendees to his Thursday MICCAI session for the full story.
More from Research
- Researchers debate GPT-6 Astra: a generalist VLA that could beat custom VLAs — YouJiacheng · 2026-09-28
- Eric Jang: now is the window to test OSS models on robotics before the gap closes — ericjang11 · 2026-09-28
- Simulating cellular automata on an 'LLM computer' to show thinking is computational — ctjlewis · 2026-09-28
- 10 NeurIPS 2026 robotics papers to watch: VLA inference and long-horizon tasks — shaohua0116 · 2026-09-28
- Poincaré conjecture fully formalized in Lean after two-week sprint — burny_tech · 2026-09-28
- GPT-6 Astra unlocks agentic real2sim: vision models as tools to turn room video into interactive sims — ericjang11 · 2026-09-28