[1912.02178] Fantastic Generalization Measures and Where to Find Them

Abstract:Generalization of deep networks has been of great interest in recent years, resulting in a number of theoretically and empirically motivated complexity measures. However, most papers proposing such measures study only a small set of models, leaving open the question of whether the conclusion drawn from those experiments would remain valid in other settings. We present the first large scale study of generalization in deep networks. We investigate more then 40 complexity measures taken from both theoretical bounds and empirical studies. We train over 10,000 convolutional networks by systematically varying commonly used hyperparameters. Hoping to uncover potentially causal relationships between each measure and generalization, we analyze carefully controlled experiments and show surprising failures of some measures as well as promising measures for further research.

Fantastic Generalization Measures and Where to Find Them Yiding Jiang∗, Behnam Neyshabur∗, Hossein Mobahi Dilip Krishnan, Samy Bengio Google {ydjiang,neyshabur,hmobahi,dilipkay,bengio}@google.com…

saved by

• Michael Li

related reading

• Understanding deep learning requires rethinking generalizationarxiv.org
• arxiv.org/pdf/1805.08522arxiv.org
• [2503.02113] Deep Learning is Not So Mysterious or Differentarxiv.org
• [1802.05296] Stronger generalization bounds for deep nets via a compression approacharxiv.org
• Deep Learning is Not So Mysterious or Differentarxiv.org
• The Scaling Hypothesis · Gwern.netgwern.net
• Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasetsmathai-iclr.github.io
• The Generalization Mystery: Sharp vs Flat Minimainference.vc
• Just Ask for Generalization | Eric Jangevjang.com
• Maybe I was too harsh on deep learning theory (three days ago) — LessWronglesswrong.com
• Scaling Laws, Carefully | Lil'Loglilianweng.github.io
• [2605.01172] A Theory of Generalization in Deep Learningarxiv.org