AlphaProof, every mention

No Priors Ep. 90 | With Google's DeepMind's AlphaProof Team Nov 14, 2024 · 27 mentions

• ▶ 0:08 Sarah Guo Today, we have Thomas Hubert, Rishi Mehta, and Laurence Sartran from DeepMind's AlphaProof team. 4 times in the scene
• ▶ 3:05 Sarah Guo They feel impossibly hard and Alphaproof had this, um, really amazing results of solving four of the six problems this year. 4 times in the scene
• ▶ 6:37 Sarah Guo Um, is there a way you can characterize like the search space for math overall, or what alpha proof is better at, um, in terms of domains and others, like types of reasoning within math? 2 times in the scene
• ▶ 9:28 Elad Gil Or what are the limitations or approaches to scaling or, you know, uh, increasing the set of problems that alpha proof can solve? 4 times in the scene
• ▶ 13:35 Thomas Hubert So if we want to take our proof all the way there, I think that's a capability we, we need to, um, either it emerges, it's possible because, you know, like,
• ▶ 17:10 Laurent Sartran So the, the core of alpha proof, um, seems to be one of the last big, uh, challenge domains where AI still had a long way to go.
• ▶ 24:57 Laurent Sartran The way alpha proof currency operates there is that it discovers its own proofs, and when they are valid, it learns from them and develop its own style, um, which has been commented upon as looking, yeah, quite, quite alien. 2 times in the scene
• ▶ 28:23 Elad Gil How many mathematicians have access right now to alpha proof, or how do you think about engaging with the mathematics community about day-to-day usage of this pretty amazing, um, set of advancements? 2 times in the scene
• ▶ 30:59 Sarah Guo Is there something that has felt alien or surprising in terms of like an alpha proof proof so far that you can talk about? 6 times in the scene
• ▶ 34:48 Sarah Guo Can we start with advice for people who are working on math today and investing time in it, given, uh, alpha proof and related models?