1 paper · 1 filter
A. Emin Orhan
Training large-scale image recognition models is computationally expensive. This raises the question of whether there might be simple ways to improve the test performance of an alr…