NeurIPS paper: Selective Disk Bispectrum reconstructs images from bispectrum coefficients alone
ninamiolane · x · 2026-11-21
A NeurIPS 2026 paper from Nina Miolane's lab (Adele Myers) introduces the Selective Disk Bispectrum (SDB).
- Extracts shapes from images without expensive registration/alignment optimization
- Offers a principled way to build group-invariant neural networks
- First algorithm to reconstruct an image from its disk bispectrum coefficients alone; SDB is the minimal subset necessary for inversion
More from Research
- Harvard chemistry polymath George Whitesides dies, leaving ~1000 papers across chemistry — CatAstro_Piyush · 2026-11-21
- a16z researcher: AI can solve Millennium math problems with Lean proofs but can't explain them — a16z · 2026-11-21
- AACR sets Data Science and AI in Oncology conference for March 2027 in San Francisco — jnkath · 2026-11-21
- Riemann proof released as blog post and 8 zips, no listed author — eigenhector · 2026-11-21
- NeurIPS Paper Derives First Algorithm to Reconstruct Images from Disk Bispectrum Alone — ninamiolane · 2026-11-21
- Impetus Grants Appeals for Philanthropic Funding to Build the Biological Data AI Cures Need — shae_mcl · 2026-11-21