Mathematical proofs, computational complexity and AI
Nancy Abigail Nuñez (Universidad Nacional Autónoma de México)
| Thu Dec 10, 19:00-20:00 (3 weeks from now) | |
Abstract: Mathematical proofs are a crucial aspect of mathematicians' work and, in many cases, finding or discovering a proof is quite challenging. Finding the proof of certain mathematical theorems, such as Fermat's Last Theorem or the Four Color Theorem, required decades or even centuries, which attests to the difficulty of establishing the truth of some mathematical statements. Nevertheless, the rapid progress of artificial intelligence currently suggests that, in the near future, it will be possible to automate the discovery of mathematical proofs; that is, that artificial intelligence will be able to prove mathematical theorems. This talk critically examines the viability of relying on artificial intelligence to fully automate the discovery of mathematical proofs, in light of certain results from computational complexity theory. Without disregarding the recent advances achieved by artificial intelligence in this area (such as the solution announced by OpenAI to the Navier–Stokes problem), this talk seeks to temper the expectations placed on this technology to fully replace human reasoning in the discovery of mathematical proofs, since the complexity of this task defies automation. The hypothesis of this talk is that, because of the complexity inherent in discovering or searching for mathematical proofs, a broad class of problems may not be amenable to automation by artificial intelligence, automated theorem provers, or proof assistants.
computational complexitycomputation and languagelogic
Audience: researchers in the topic
Series comments: Description: Seminar on all areas of logic
| Organizer: | Wesley Calvert* |
| *contact for this listing |
