University of Cambridge > Talks.cam > CamPoS (Cambridge Philosophy of Science) seminar > What kind of models are deep learning algorithms?

What kind of models are deep learning algorithms?

Add to your list(s) Download to your calendar using vCal

  • UserLena Zuchowski (University of Bristol)
  • ClockWednesday 03 March 2021, 13:00-14:30
  • HouseZoom.

If you have a question about this talk, please contact Matt Farr.

I will introduce a novel conceptual framework for the analysis of scientific modelling. The framework will be used to distinguish and comparatively analyse three different ways of model construction: vertical from covering theory and empirical knowledge about a given target system; horizontal through the systematic variation or transfer of existing models; and diagonal through a combination of vertical and horizontal construction steps. I will then apply this framework to analyse the construction of deep learning algorithms and will argue that they can be interpreted as the automated, vertical, bottom-up construction of a sequence of scientific models. Furthermore, I will maintain that the practice of transfer learning can be interpreted as horizontal model construction.

This talk is part of the CamPoS (Cambridge Philosophy of Science) seminar series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2021 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity