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

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

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

Statistics > Machine Learning

Title: Self-Certifying Classification by Linearized Deep Assignment

Abstract: We propose a novel class of deep stochastic predictors for classifying metric data on graphs within the PAC-Bayes risk certification paradigm. Classifiers are realized as linearly parametrized deep assignment flows with random initial conditions. Building on the recent PAC-Bayes literature and data-dependent priors, this approach enables (i) to use risk bounds as training objectives for learning posterior distributions on the hypothesis space and (ii) to compute tight out-of-sample risk certificates of randomized classifiers more efficiently than related work. Comparison with empirical test set errors illustrates the performance and practicality of this self-certifying classification method.
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)
Cite as: arXiv:2201.11162 [stat.ML]
  (or arXiv:2201.11162v2 [stat.ML] for this version)

Submission history

From: Bastian Boll [view email]
[v1] Wed, 26 Jan 2022 19:59:14 GMT (129kb,D)
[v2] Fri, 18 Feb 2022 10:22:01 GMT (115kb,D)

Link back to: arXiv, form interface, contact.