pith. sign in

arxiv: 1908.07380 · v5 · pith:PQMV4KFBnew · submitted 2019-08-19 · 💻 cs.LG · stat.ML

PAC-Bayes with Backprop

classification 💻 cs.LG stat.ML
keywords pac-bayesmnistbackpropboundboundsdataderivedneural
0
0 comments X
read the original abstract

We explore the family of methods "PAC-Bayes with Backprop" (PBB) to train probabilistic neural networks by minimizing PAC-Bayes bounds. We present two training objectives, one derived from a previously known PAC-Bayes bound, and a second one derived from a novel PAC-Bayes bound. Both training objectives are evaluated on MNIST and on various UCI data sets. Our experiments show two striking observations: we obtain competitive test set error estimates (~1.4% on MNIST) and at the same time we compute non-vacuous bounds with much tighter values (~2.3% on MNIST) than previous results. These observations suggest that neural nets trained by PBB may lead to self-bounding learning, where the available data can be used to simultaneously learn a predictor and certify its risk, with no need to follow a data-splitting protocol.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. From Privacy to Generalization: Linear Max-Information Bounds for DP-SGD

    cs.LG 2026-05 unverdicted novelty 6.0

    Proves linear max-information bound for DP-SGD and derives explicit PAC-Bayes generalization bounds with learnable prior and hyperparameter-controlled complexity term.

  2. Federated Learning with Nonvacuous Generalisation Bounds

    cs.LG 2023-10 unverdicted novelty 6.0

    Federated learning trains private local randomised predictors whose aggregation yields a global predictor with nonvacuous PAC-Bayesian generalisation bounds and near-centralized accuracy.

  3. Margin-Adaptive Confidence Ranking for Reliable LLM Judgement

    cs.LG 2026-05 unverdicted novelty 5.0

    Introduces a margin-adaptive confidence ranking method that learns an estimator from simulated diversity and derives margin-dependent generalization bounds for use in fixed-sequence testing of LLM-human agreement.

  4. Margin-Adaptive Confidence Ranking for Reliable LLM Judgement

    cs.LG 2026-05 unverdicted novelty 4.0

    Develops a margin-adaptive learned confidence estimator for LLMs with generalization guarantees to improve agreement rates with human judgments over heuristic baselines.