We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Towards explaining the generalization gap in neural networks using topological data analysis

Abstract: Understanding how neural networks generalize on unseen data is crucial for designing more robust and reliable models. In this paper, we study the generalization gap of neural networks using methods from topological data analysis. For this purpose, we compute homological persistence diagrams of weighted graphs constructed from neuron activation correlations after a training phase, aiming to capture patterns that are linked to the generalization capacity of the network. We compare the usefulness of different numerical summaries from persistence diagrams and show that a combination of some of them can accurately predict and partially explain the generalization gap without the need of a test set. Evaluation on two computer vision recognition tasks (CIFAR10 and SVHN) shows competitive generalization gap prediction when compared against state-of-the-art methods.
Comments: 20 pages, 7 figures
Subjects: Machine Learning (cs.LG); Algebraic Topology (math.AT)
MSC classes: 55N31, 68T07
ACM classes: I.2.6
Cite as: arXiv:2203.12330 [cs.LG]
  (or arXiv:2203.12330v1 [cs.LG] for this version)

Submission history

From: Carles Casacuberta [view email]
[v1] Wed, 23 Mar 2022 11:15:36 GMT (870kb,D)

Link back to: arXiv, form interface, contact.