VibeMathedMath problems solved by AI

Online Shadow Tomography Matching the Classical Bounds

Online Shadow Tomography with logm\log m dependence, while retaining poly(log(d)/ϵ)\mathrm{poly}(\log(d)/\epsilon) dependence. Also, matching the best classical bounds for Adaptive Data Analysis

Result
Proved
Status
Resolved
AI contribution
AI-discovered
Method
Argument
Field
Quantum information & computing
Posed by
Scott Aaronson
Year posed
2016
Years open
10y
Solved
2026-07-31
Model
ChatGPT 5.6-Sol Pro
Vendor
OpenAI
Collaborators
Sitan Chen, Ryan O'Donnell, Angelos Pelecanos, John Wright
Verification
Unreviewed
Publication
Preprint
Significance
Disclosed cost
Wikipedia
No dedicated article

What the AI did

Discovered the main proof

Verification

Unreviewed preprint. The authors state they studied, refined and verified the model's ideas themselves and take full responsibility for every claim, proof and citation; that is the authors checking their own work, so it stays Unreviewed until someone independent looks. Recorded as Resolved rather than Partial because the stated target - matching the classical Adaptive Data Analysis rates - is fully achieved by Theorems 1.2 and 1.3. What remains open is whether those rates are optimal, which was never the question this entry records.

Source

Submitted by WittyFerret553 on

Changelog3 changes
  • Rasmus Lindahlapproved this entry
  • Rasmus Lindahlchanged Status from partial to resolved, also Model, Verification note
  • WittyFerret553submitted this entry

Discussion