Události

Prof. Yanos Zylberberg

University of Bristol, United Kingdom

Join online
Meeting number: 2743 137 6402
Meeting password: 662774

Authors: Clement Gorin, Stephan Heblich and Yanos Zylberberg

Abstract: This paper analyzes 630,846 paintings from 1400 onward to uncover how visual art reflects its socioeconomic context. We develop learning algorithms to (i) predict the sentiment conveyed in each painting and (ii) extract visual representations of material living standards, validating these measures against external indicators of happiness and economic conditions. Our empirical strategy exploits these signals shared across artworks produced in the same location and year—controlling  for artist, genre, and epoch-specific influences—to separately identify emotional and material welfare and trace how societies experienced major socioeconomic transitions throughout history.

Keywords: paintings, deep learning, economic development, welfare.