Prior Distribution and Model Confidence
We study how the training data distribution affects confidence and performance in image classification models. We introduce Embedding Density, a model-agnostic framework that estimates prediction confidence by measuring the distance of test samples from the training distribution in embedding space, without requiring retraining. By filtering low-density (low-confidence) predictions, our method significantly improves classification accuracy. We evaluate Embedding Density across multiple architectures and compare it with state-of-the-art out-of-distribution (OOD) detection methods. The proposed approach is potentially generalizable beyond computer vision.
Code (0)
등록된 구현이 없습니다.
Tasks
Image ClassificationSimilar Papers 제목 키워드 기반
On Prior Confidence and Belief Updating
We experimentally investigate how confidence over multiple priors affects belief updating. Theory predicts that the average Bayesian posterior is unaffected by confidence over multiple priors if average priors are the sa…
Believe Your Model: Distribution-Guided Confidence Calibration
Large Reasoning Models have demonstrated remarkable performance with the advancement of test-time scaling techniques, which enhances prediction accuracy by generating multiple candidate responses and selecting the most r…
Answer SelectionAsymptotically Log-Optimal Bayes-Assisted Confidence Sequences for Bounded Means
Confidence sequences based on test martingales provide time-uniform uncertainty quantification for the mean of bounded IID observations without parametric distributional assumptions. Their practical efficiency, however, …
Combining Priors with Experience: Confidence Calibration Based on Binomial Process Modeling
Confidence calibration of classification models is a technique to estimate the true posterior probability of the predicted class, which is critical for ensuring reliable decision-making in practical applications. Existin…
Can a Confident Prior Replace a Cold Posterior?
Benchmark datasets used for image classification tend to have very low levels of label noise. When Bayesian neural networks are trained on these datasets, they often underfit, misrepresenting the aleatoric uncertainty of…
image-classificationImage Classification