How a paper debunking NeuroAI's grid-cell myth shaped a researcher's path
RylanSchaeffer · x · 2026-10-20
Schaeffer's first 'valuable' paper challenged a prevailing NeuroAI myth: that training networks on path integration would yield the Nobel-winning grid-cell representations found in mammalian brains. The attached figure shows grid-like responses require highly specific target encodings — Cartesian, polar and Gaussian readouts learn the task but not grid cells, while a tuned Difference-of-Softmaxes readout does. He shares the paper's backstory for the first time, quoting Steve Jobs on following your heart. The quoted tweet adds three papers on streaming variational inference for Bayesian nonparametrics, with the lesson that sophisticated methods that don't scale lose to simple ones that do.
More from Research
- RFD4-Proteina's 'underrated' docs detail partial diffusion for protein redesign — AllThingsApx · 2026-10-23
- MPI-Sec PhD Thesis on Vulnerability Detection Benchmarks Includes Pre-LLM Classic "Top Score on the Wrong Exam" — mboehme_ · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- New paper makes 9 falsifiable predictions toward a major AI-enabled terror attack by 2027-2028 — StephenLCasper · 2026-10-23
- HumanoidPose recovers humanoid robots' full-body joint states from RGB video alone — cholz · 2026-10-23
- Sophont's Holon-S1 predicts genes and biomarkers from routine pathology slides — iScienceLuvr · 2026-10-23