What is Flagged in Uncertainty Quantification? Latent Density Models for Uncertainty Categorization
Explore this paper's citation graph
Summary
This work proposes a framework for categorizing uncertain examples flagged by UQ methods in classification tasks and introduces the confusion density matrix, a kernel-based approximation of the misclassification density, to categorize suspicious examples identified by a given uncertainty method into three classes.
- Type
- preprint
- Published
- 2022-07-11
- Cited by
- 6
- References
- 56
- Access
- Open access
- OpenAlex
- https://openalex.org/W4285429072
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:250450892
Keywords
Uncertainty quantification, Computer science, Categorization, Benchmark (surveying), Machine learning
References
- Conformal Prediction for Reliable Machine Learning: Theory, Adaptations and Applications
- A reliable data-based bandwidth selection method for kernel density estimation
- Algorithmic Learning in a Random World
- Practical Variational Inference for Neural Networks
- Weight Uncertainty in Neural Network
- Visualizing Data using t-SNE
- Isolation Forest
- On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learning
- Finding label noise examples in large scale datasets
- Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors
- Empirical Study of Easy and Hard Examples in CNN Training
- Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
- Likelihood Ratios for Out-of-Distribution Detection
- What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
- A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
- Quality of Uncertainty Quantification for Bayesian Neural Network Inference
- Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention
- Predictive Uncertainty Estimation via Prior Networks
- Density estimation in representation space to predict model uncertainty
- Can Deep Learning Predict Risky Retail Investors? A Case Study in Financial Risk Behavior Forecasting
Cited by
- RORL: Robust Offline Reinforcement Learning via Conservative Smoothing
- A Survey of Confidence Estimation and Calibration in Large Language Models
- Disentangling Uncertainties by Learning Compressed Data Representation
- Uncertainty Reasoning with Photonic Bayesian Machines
- Finer Disentanglement of Aleatoric Uncertainty Can Accelerate Chemical Histopathology Imaging
- Uncertainty-Aware Systems for Human-AI Collaboration
Related papers
- Two categorization patterns in idiom semantics
- On the Reasons for Cognitive Differences During Categorization
- Is one object enough? Diagnosticity of single objects for fast scene categorization
- Visual search and the collapse of categorization.
- Categorization by elimination : a fast and frugal approach tocategorization
- The future of infant categorization research: a process-oriented approach.
- The Role of Comparison in Perschoolers' Categorization
- Cultural Influences on Categorization Processes