mlsanityData to 7 Oct 2026

Identifying Generalization Properties in Neural Networks

arXiv 1809.07402 ↗First posted 19 Sep 2018cs.LG stat.MLcited by 15

Not on the map. Why?

Abstract (arXiv)

While it has not yet been proven, empirical evidence suggests that model generalization is related to local properties of the optima which can be described via the Hessian. We connect model generalization with the local property of a solution under the PAC-Bayes paradigm. In particular, we prove that model generalization ability is related to the Hessian, the higher-order "smoothness" terms characterized by the Lipschitz constant of the Hessian, and the scales of the parameters. Guided by the proof, we propose a metric to score the generalization capability of the model, as well as an algorithm that optimizes the perturbed model accordingly.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). First posted on arXiv 19 Sep 2018.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 5; 2022: 4; 2023: 1; 2024: 4; 2025: 1.

15 papers.

Sort:

By area:

  1. 2023Deep Model Fusion: A Surveycited by 89
  2. 2022Anticorrelated Noise Injection for Improved Generalizationcited by 41
  3. 2022ALLSH: Active Learning Guided by Local Sensitivity and Hardnesscited by 23
  4. 2021Effective dimension of machine learning modelscited by 14
  5. 2025Self-Destructive Language Modelcited by 10
  6. 2021Optimization Variance: Exploring Generalization Properties of DNNscited by 4
  7. 2021Minimum sharpness: Scale-invariant parameter-robustness of neural networkscited by 4
  8. 2021BN-invariant sharpness regularizes the training model to better generalizationcited by 3
  9. 2022Beyond accuracy: generalization properties of bio-plausible temporal credit assignment rulescited by 3
  10. 2024A Method for Enhancing Generalization of Adam by Multiple Integrationscited by 2
  11. 2022Attack-Agnostic Adversarial Detectioncited by 1
  12. 2024Dimension-free uniform concentration bound for logistic regressioncited by 1
3 more
  1. 2021LRTuner: A Learning Rate Tuner for Deep Neural Networkscited by 0
  2. 2024DeepKnowledge: Generalisation-Driven Deep Learning Testingcited by 0
  3. 2024IKUN: Initialization to Keep snn training and generalization great with sUrrogate-stable variaNcecited by 0