EPFL quantum CNN learns digits from 10 samples where a 45-param classical CNN stays at chance
PlisSergey · x · 2026-09-23
- EPFL/FHNW researchers (Anthony, Burov, Piro, Dal Peraro, Javerzac) published a paper showing a quantum convolutional neural network with mid-circuit measurement learns digit classification from just 10 samples, while a comparably accurate 45-parameter classical CNN remains at chance.
- They argue data encoding, not optimization, is now the ceiling for quantum ML in drug discovery.
- The poster uses this to promote their own service with heavy marketing claims; the substantive value lies in the quoted paper.
More from Research
- Yale PhD student open-sources his paper figure scripts, packaged as a Skill for Claude Code and Cursor — burny_tech · 2026-09-23
- AI models now match superforecasters on ForecastBench; rematch set for October — burny_tech · 2026-09-23
- Dev uses Opus 5.5 with Lean to formally verify Claude Agent SDK, yielding 16 bug-fix PRs — bcherny · 2026-09-23
- Mathematicians, not just LLMs, made AI's math breakthroughs possible, scholars argue — tak3sh8 · 2026-09-23
- AI-enabled drug discovery cuts discovery time by 15-80%, McKinsey research finds — menhguin · 2026-09-23
- Gemini training details dissected: groupwise reward redistribution to fight reward hacking — nrehiew_ · 2026-09-23