REVIEW 1 cited by
Persistent Homology Captures the Generalization of Neural Networks Without A Validation Set
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The training of neural networks is usually monitored with a validation (holdout) set to estimate the generalization of the model. This is done instead of measuring intrinsic properties of the model to determine whether it is learning appropriately. In this work, we suggest studying the training of neural networks with Algebraic Topology, specifically Persistent Homology (PH). Using simplicial complex representations of neural networks, we study the PH diagram distance evolution on the neural network learning process with different architectures and several datasets. Results show that the PH diagram distance between consecutive neural network states correlates with the validation accuracy, implying that the generalization error of a neural network could be intrinsically estimated without any holdout set.
Forward citations
Cited by 1 Pith paper
-
A Quotient Homology Theory of Representation in Neural Networks
For ReLU networks, the homology of the output representation is isomorphic to the homology of the input manifold quotiented by the network's overlap decomposition, when polyhedron-manifold intersections are convex.
Discussion (0). Continue with ORCID to comment.