Overcoming the Boundaries of Artificial Intelligence: A Mathematical Approach
Gitta Kutyniok
Abstract: Classical approaches of artificial intelligence typically employ digital hardware. However, it turns out that such computing platforms impose serious restrictions to AI-based algorithms in terms of computability, reliability, legal requirements, and energy requirements.
In this third lecture, we will first discuss current mathematical limitations of artificial intelligence imposed by digital hardware modeled as a Turing machine. We will then show how those boundaries can be overcome by embracing analog computing approaches, modeled by the Blum-Shub-Smale machine. This will reveal the tremendous importance of novel computing hardware such as neuromorphic hardware for future AI computing. Finally, we will discuss mathematical aspects of spiking neural networks, which mimic natural neural networks much closer than classical artificial neural networks and are perfectly adapted to neuromorphic hardware.
Mathematics
Audience: researchers in the topic
UCLA distinguished lecture series
Series comments: Description: Lectures by distinguished mathematicians, hosted at UCLA
Every year, the Distinguished Lecture Series (DLS) brings two to four eminent mathematicians to UCLA for a week or more to give a lecture series on their field, and to meet with faculty and graduate students.
The first lecture of each series is aimed at a general mathematical audience, and offers a rare opportunity to see the state of an area of mathematics from the perspective of one of its leaders. The remaining lectures in the series are usually more advanced, concerning recent developments in the area.
Organizer: | Terence Tao* |
*contact for this listing |