تقرير
Learning Expressive Priors for Generalization and Uncertainty Estimation in Neural Networks
العنوان: | Learning Expressive Priors for Generalization and Uncertainty Estimation in Neural Networks |
---|---|
المؤلفون: | Schnaus, Dominik, Lee, Jongseok, Cremers, Daniel, Triebel, Rudolph |
سنة النشر: | 2023 |
المجموعة: | Computer Science Statistics |
مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Statistics - Machine Learning |
الوصف: | In this work, we propose a novel prior learning method for advancing generalization and uncertainty estimation in deep neural networks. The key idea is to exploit scalable and structured posteriors of neural networks as informative priors with generalization guarantees. Our learned priors provide expressive probabilistic representations at large scale, like Bayesian counterparts of pre-trained models on ImageNet, and further produce non-vacuous generalization bounds. We also extend this idea to a continual learning framework, where the favorable properties of our priors are desirable. Major enablers are our technical contributions: (1) the sums-of-Kronecker-product computations, and (2) the derivations and optimizations of tractable objectives that lead to improved generalization bounds. Empirically, we exhaustively show the effectiveness of this method for uncertainty estimation and generalization. Comment: Accepted to ICML 2023 |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2307.07753 |
رقم الأكسشن: | edsarx.2307.07753 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |