Could AI push approximation algorithms beyond current limits on TSP?
chaitjo · x · 2026-08-04
The post asks whether AI systems like Astra and Fable could make a real advance on theoretical approximation limits for problems such as the Traveling Salesman Problem.
It links to a Quanta Magazine profile of Shayan Oveis Gharan, who won the 2026 Abacus Medal for using tools from across mathematics to improve algorithms. The article frames his work as progress on hard problems coming from unexpected detours and cross-disciplinary methods.
More from AGI Musings
- Andrew Yang: AI's Impact on Work is a Step Change, Not Linear — Olivier__OG · 2026-08-04
- LLMs Show High Intelligence Without Consciousness: Are They Separate? — GateSpiritual5717 · 2026-08-04
- OpenAI’s Wojciech Zaremba says AI safety may need a fire-like resilience stack — richie9830 · 2026-08-04
- Andrew Chen says AI product-market fit now follows the frontier — andrewchen · 2026-08-04
- A slowdown letter may mainly restrain frontier labs and help others catch up — 12exyz · 2026-08-04
- If reality is a simulation, the post argues the far future should be discounted more heavily — EigenGender · 2026-08-04