Emergence of Concepts in DNNs?

11/11/2022
by   Tim Räz, et al.
0

The present paper reviews and discusses work from computer science that proposes to identify concepts in internal representations (hidden layers) of DNNs. It is examined, first, how existing methods actually identify concepts that are supposedly represented in DNNs. Second, it is discussed how conceptual spaces – sets of concepts in internal representations – are shaped by a tradeoff between predictive accuracy and compression. These issues are critically examined by drawing on philosophy. While there is evidence that DNNs able to represent non-trivial inferential relations between concepts, our ability to identify concepts is severely limited.

READ FULL TEXT
research
01/11/2018

Formalized Conceptual Spaces with a Geometric Representation of Correlations

The highly influential framework of conceptual spaces provides a geometr...
research
06/20/2017

A Thorough Formalization of Conceptual Spaces

The highly influential framework of conceptual spaces provides a geometr...
research
04/06/2018

Formal Ways for Measuring Relations between Concepts in Conceptual Spaces

The highly influential framework of conceptual spaces provides a geometr...
research
07/07/2017

Measuring Relations Between Concepts In Conceptual Spaces

The highly influential framework of conceptual spaces provides a geometr...
research
03/21/2022

The Conceptual VAE

In this report we present a new model of concepts, based on the framewor...
research
05/03/2018

Learning Conceptual Space Representations of Interrelated Concepts

Several recently proposed methods aim to learn conceptual space represen...
research
09/01/2021

A model for discovering 'containment' relations

Rapid developments in the fields of learning and object recognition have...

Please sign up or login with your details

Forgot password? Click here to reset