ETH-FDS Stiefel lectures

On this website you can find information about upcoming and past lectures.

Modal title

Modal content

Autumn Semester 2019

Date / Time

Speaker

Title

Location

8 November 2019
17:15-18:15

David Donoho
Stanford University

Event Details

ETH-FDS Stiefel Lectures

Title	Deepnet Spectra and the two cultures of data science
Speaker, Affiliation	David Donoho, Stanford University
Date, Time	8 November 2019, 17:15-18:15
Location	HG F 1
Abstract	Machine learning became a remarkable media story of the 2010s largely owing to its ability to focus researcher energy on attacking prediction challenges like ImageNet. Media extrapolation of complete transformation of human existence has (predictably) ensued. Unfortunately machine learning has a troubled relationship with understanding the foundation of its achievements well enough to face demanding real world requirements outside the challenge setting. For example, its literature is admittedly corrupted by anti intellectual and anti scholarly tendencies. It is beyond irresponsible to build a revolutionary transformation on such a shaky pseudo-foundation. In contrast, more traditional subdisciplines of data science like numerical linear algebra, applied probability, and theoretical statistics provide time-tested tools for designing reliable processes with understandable performance. Moreover, positive improvements in human well being have repeatedly been constructed using these foundations. To illustrate these points we will review a recent boomlet in the ML literature in the study of eigenvalues of deepnet Hessians. A variety of intriguing patterns in eigenvalues were observed and speculated about in ML conference papers. We describe work of Vardan Papyan showing that the traditional subdisciplines, properly deployed, can offer insights about these objects that ML researchers had been seeking.
Assets	Video David Donoho - Stiefel Lecture 2019file_download

Deepnet Spectra and the two cultures of data science read_more

HG F 1

Note: if you want you can subscribe to the iCal/ics Calender.

Archive: SS 24 AS 23 AS 22 SS 22 AS 19