Timezone: »

NoFADE: Analyzing Diminishing Returns on CO2 Investment
Andre Fu · Justin Tran · Andy Xie · Jonathan Spraggett · Elisa Ding · Chang-Won Lee · Kanav Singla · Mahdi Hosseini · Konstantinos N Plataniotis
Event URL: https://www.climatechange.ai/papers/neurips2021/46 »

Climate change continues to be a pressing issue that currently affects society at-large. It is important that we as a society, including the Computer Vision (CV) community take steps to limit our impact on the environment. In this paper, we (a) analyze the effect of diminishing returns on CV methods, and (b) propose a \textit{``NoFADE''}: a novel entropy-based metric to quantify model--dataset--complexity relationships. We show that some CV tasks are reaching saturation, while others are almost fully saturated. In this light, NoFADE allows the CV community to compare models and datasets on a similar basis, establishing an agnostic platform.

Author Information

Andre Fu (University of Toronto)
Justin Tran (University of Toronto)
Andy Xie (University of Toronto)
Jonathan Spraggett (University of Toronto)
Elisa Ding (University of Toronto)
Chang-Won Lee (University of Toronto)
Kanav Singla (University of Toronto )
Mahdi Hosseini (University of New Brunswick)
Konstantinos N Plataniotis (UofT)

More from the Same Authors