paper-with-me

Papers

Isotropy Maximization Loss and Entropic Score: Accurate, Fast, Efficient, Scalable, and Turnkey Neural Networks Out-of-Distribution Detection Based on The Principle of Maximum Entropy

2019-08-15 · David Macêdo, Tsang Ing Ren, Cleber Zanchettin, Adriano L. I. Oliveira, Teresa Ludermir

Current out-of-distribution (OOD) detection approaches require cumbersome procedures that add undesired side effects to the solution. In this paper, we argue that the low OOD detection performance of neural networks is due to cross-entropy SoftMax loss anisotropy and extreme propensity to produce low entropy (high confidence) posterior probability distributions in frontal disagreement with the Principle of Maximum Entropy. Consequently, we propose IsoMax, a loss that is isotropic (distance-based) and produces high entropy (low confidence) posterior probability distributions despite still relying on cross-entropy minimization. Additionally, we propose a speedy Entropic Score for OOD detection. IsoMax loss works as a seamless SoftMax loss drop-in replacement that keeps the overall solution accurate, fast, efficient, scalable, and turnkey. Our experiments indeed confirmed that neural networks OOD detection performance may be extremely improved without relying on techniques such as adversarial training or validation, data augmentation, ensembles methods, generative approaches, model architectural changes, metric learning, or additional classifiers or regressions. The results also showed that our straightforward approach is competitive against state-of-the-art solutions besides avoiding previous methods undesired drawbacks.

📄 PDF Abstract BibTeX arXiv:1908.05569

Code (1)

dlmacedo/entropic-out-of-distribution-detection 공식 구현 pytorch

Tasks

Data AugmentationMetric LearningOut-of-Distribution DetectionOut of Distribution (OOD) Detection

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Enhanced Isotropy Maximization Loss: Seamless and High-Performance Out-of-Distribution Detection Simply Replacing the SoftMax Loss

2021-05-30 · David Macêdo, Teresa Ludermir

Current out-of-distribution detection approaches usually present special requirements (e.g., collecting outlier data and hyperparameter validation) and produce side effects (e.g., classification accuracy drop and slow/in…

Out-of-Distribution Detection

Partial local entropy and anisotropy in deep weight spaces

2020-07-17 · Daniele Musso

We refine a recently-proposed class of local entropic loss functions by restricting the smoothening regularization to only a subset of weights. The new loss functions are referred to as partial local entropies. They can …

image-classificationImage Classification

Improving Entropic Out-of-Distribution Detection using Isometric Distances and the Minimum Distance Score

2021-05-21 · NeurIPS 2021 12 · David Macêdo, Teresa Ludermir

Current out-of-distribution detection approaches usually present special requirements (e.g., collecting outlier data and hyperparameter validation) and produce side effects (classification accuracy drop and slow/ineffici…

Out-of-Distribution Detection

IsoScore: Measuring the Uniformity of Embedding Space Utilization

2021-08-16 · Findings (ACL) 2022 5 · William Rudman, Nate Gillman, Taylor Rayne, Carsten Eickhoff

The recent success of distributed word representations has led to an increased interest in analyzing the properties of their spatial distribution. Several studies have suggested that contextualized word embedding models …

Word Embeddings

Tsallis Reinforcement Learning: A Unified Framework for Maximum Entropy Reinforcement Learning

2019-01-31 · Kyungjae Lee, Sungyub Kim, Sungbin Lim, Sungjoon Choi 외

In this paper, we present a new class of Markov decision processes (MDPs), called Tsallis MDPs, with Tsallis entropy maximization, which generalizes existing maximum entropy reinforcement learning (RL). A Tsallis MDP pro…

MuJoCoreinforcement-learningReinforcement LearningReinforcement Learning (RL)