A Boo(n) for Evaluating Architecture Performance

Ondrej Bajgar, Rudolf Kadlec, and Jan Kleindienst

Feb 15, 2018 (modified: Jul 23, 2018) ICLR 2018 Conference Blind Submission readers: everyone Show Bibtex
  • Abstract: We point out important problems with the common practice of using the best single model performance for comparing deep learning architectures, and we propose a method that corrects these flaws. Each time a model is trained, one gets a different result due to random factors in the training process, which include random parameter initialization and random data shuffling. Reporting the best single model performance does not appropriately address this stochasticity. We propose a normalized expected best-out-of-n performance (Boo_n) as a way to correct these problems.
  • TL;DR: We point out important problems with the common practice of using the best single model performance for comparing deep learning architectures, and we propose a method that corrects these flaws.
  • Keywords: evaluation, methodology
0 Replies

Loading