Troubling Trends in Machine Learning Scholarship

Spurious Theorems

Spurious theorems are common culprits, inserted into papers to lend authoritativeness to empirical results, even when the theorem’s conclusions do not actually support the main claims of the paper.

This is a perfect description of a lot of theorems in machine learning papers (and, for that matter, statistics papers too). As a recent example, Theorem 2 in (Arora, Cohen, and Hazan ⊕2018Arora, Sanjeev, Nadav Cohen, and Elad Hazan. 2018. “On the optimization of deep networks: Implicit acceleration by overparameterization.” In 35th International Conference on Machine Learning, Icml 2018, 372–89. Institute for Advanced Studies, Princeton, United States.) is border-line spurious.

Backlinks

• [[on-the-optimization-of-deep-networks]]
  • this proof felt a little spurious [[troubling-trends-in-machine-learning-scholarship]]