paper-with-me

홈 › Papers

Preventing Arbitrarily High Confidence on Far-Away Data in Point-Estimated Discriminative Neural Networks

2023-11-07 · Ahmad Rashid, Serena Hacker, Guojun Zhang, Agustinus Kristiadi, Pascal Poupart

Discriminatively trained, deterministic neural networks are the de facto choice for classification problems. However, even though they achieve state-of-the-art results on in-domain test sets, they tend to be overconfident on out-of-distribution (OOD) data. For instance, ReLU networks - a popular class of neural network architectures - have been shown to almost always yield high confidence predictions when the test data are far away from the training set, even when they are trained with OOD data. We overcome this problem by adding a term to the output of the neural network that corresponds to the logit of an extra class, that we design to dominate the logits of the original classes as we move away from the training data.This technique provably prevents arbitrarily high confidence on far-away test data while maintaining a simple discriminative point-estimate training. Evaluation on various benchmarks demonstrates strong performance against competitive baselines on both far-away and realistic OOD data.

📄 PDF Abstract BibTeX arXiv:2311.03683

Code (1)

serenahacker/preload 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Towards neural networks that provably know when they don't know

2019-09-26 · ICLR 2020 1 · Alexander Meinke, Matthias Hein

It has recently been shown that ReLU networks produce arbitrarily over-confident predictions far away from the training data. Thus, ReLU networks do not know when they don't know. However, this is a highly important prop…

Out-of-Distribution Detection

Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem

2018-12-13 · CVPR 2019 6 · Matthias Hein, Maksym Andriushchenko, Julian Bitterwolf

Classifiers used in the wild, in particular for safety-critical systems, should not only have good generalization properties but also should know when they don't know, in particular make low confidence predictions far aw…

General Classification

Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks

2020-02-24 · ICML 2020 1 · Agustinus Kristiadi, Matthias Hein, Philipp Hennig

The point estimates of ReLU classification networks---arguably the most widely used neural network architecture---have been shown to yield arbitrarily high confidence far away from the training data. This architecture, i…

Bayesian Inference

Analysis of Confident-Classifiers for Out-of-distribution Detection

2019-04-27 · Sachin Vernekar, Ashish Gaurav, Taylor Denouden, Buu Phan 외

Discriminatively trained neural classifiers can be trusted, only when the input data comes from the training distribution (in-distribution). Therefore, detecting out-of-distribution (OOD) samples is very important to avo…

General Classificationimage-classificationImage ClassificationOut-of-Distribution Detection+1

Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfaces

2025-09-29 · Huanshu Zhang, Lei Kang, Sawyer D. Campbell, Douglas H. Werner arxiv

Traditional metasurface design is limited by the computational cost of full-wave simulations, preventing thorough exploration of complex configurations. Data-driven approaches have emerged as a solution to this bottlenec…