all AI news
Immiscible Color Flows in Optimal Transport Networks for Image Classification. (arXiv:2205.02938v1 [cs.CV])
cs.LG updates on arXiv.org arxiv.org
In classification tasks, it is crucial to meaningfully exploit information
contained in data. Here, we propose a physics-inspired dynamical system that
adapts Optimal Transport principles to effectively leverage color distributions
of images. Our dynamics regulates immiscible fluxes of colors traveling on a
network built from images. Instead of aggregating colors together, it treats
them as different commodities that interact with a shared capacity on edges.
Our method outperforms competitor algorithms on image classification tasks in
datasets where color information matters.