[Yang-Hui He] Presentation to the Royal Institution on AI and Mathematics

Machine Learning


Available on YouTube [Yang-Hui He] gift to royal institution About Mathematics: The Rise of the Machines.

In this one-hour presentation, [Yang-Hui He] Learn how AI is driving advances in pure mathematics. He says that AI is now changing the very way mathematics is done. He is part of a community of hundreds of mathematicians pursuing the use of AI for research purposes.

[Yang-Hui He] He traces the origin of the term “artificial intelligence” to a research proposal from J. McCarthy, ML Minsky, N. Rochester, and CE Shannon dated August 31, 1955. He states that his belief has become that “connectivism brings emergence” and explains what that means, then follows with the universal approximation theorem.

He goes on to list some of the key moments in AI. Descartes’ Beitmachine, 1617. Lovelace’s Conjecture, 1842. Turing test, 1949. Dartmouth Conference, 1956. Rosenblatt’s Perceptron, 1957. Hopfield’s Network, 1982. Hinton’s Boltzmann Machine, 1984. IBM’s Deep Blue, 1997. and DeepMind’s AlphaGo, 2012.

He goes on to navel-gaze what mathematics is and what artificial intelligence is. He thinks of the way we do mathematics as bottom-up, top-down, or meta-mathematics. He mentions one of his early papers on the subject Machine Learning in String Landscapes (PDF) and his book Calabi Yau Landscapes: Machine Learning from Geometry to Physics and Machine Learning in Pure Mathematics and Theoretical Physics.

He goes on to discuss Mathlib and the Xena project. He will discuss Machine-Assisted Proof (PDF) by Terence Tao and talk more about the history of mathematics, especially experimental mathematics. Overall, it’s a very interesting talk, so please enjoy it if you have time.

In conclusion, has AI resolved the major open conjectures? No. Is AI starting to help advance mathematical discovery? Yes. Has AI changed speakers’ daily research routines? Yes or no.

If you’re interested in more fun math articles, be sure to check out Digital Paint Mixing Was Greatly Improved in 1930s Mathematics, Painted Over But Not Forgotten: Recovering Lost Paintings with Radiation and Mathematics.



Source link