Calibrated Decomposition of Aleatoric and Epistemic Uncertainty in Deep Features for Inference-Time Adaptation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kumar, Divake, Poggi, Patrick, Tayebati, Sina, Naik, Devashri, Ahuja, Nilesh, Trivedi, Amit Ranjan
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914160058564608
author Kumar, Divake
Poggi, Patrick
Tayebati, Sina
Naik, Devashri
Ahuja, Nilesh
Trivedi, Amit Ranjan
author_facet Kumar, Divake
Poggi, Patrick
Tayebati, Sina
Naik, Devashri
Ahuja, Nilesh
Trivedi, Amit Ranjan
contents Most estimators collapse all uncertainty modes into a single confidence score, preventing reliable reasoning about when to allocate more compute or adjust inference. We introduce Uncertainty-Guided Inference-Time Selection, a lightweight inference time framework that disentangles aleatoric (data-driven) and epistemic (model-driven) uncertainty directly in deep feature space. Aleatoric uncertainty is estimated using a regularized global density model, while epistemic uncertainty is formed from three complementary components that capture local support deficiency, manifold spectral collapse, and cross-layer feature inconsistency. These components are empirically orthogonal and require no sampling, no ensembling, and no additional forward passes. We integrate the decomposed uncertainty into a distribution free conformal calibration procedure that yields significantly tighter prediction intervals at matched coverage. Using these components for uncertainty guided adaptive model selection reduces compute by approximately 60 percent on MOT17 with negligible accuracy loss, enabling practical self regulating visual inference. Additionally, our ablation results show that the proposed orthogonal uncertainty decomposition consistently yields higher computational savings across all MOT17 sequences, improving margins by 13.6 percentage points over the total-uncertainty baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12389
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Calibrated Decomposition of Aleatoric and Epistemic Uncertainty in Deep Features for Inference-Time Adaptation
Kumar, Divake
Poggi, Patrick
Tayebati, Sina
Naik, Devashri
Ahuja, Nilesh
Trivedi, Amit Ranjan
Computer Vision and Pattern Recognition
Machine Learning
Most estimators collapse all uncertainty modes into a single confidence score, preventing reliable reasoning about when to allocate more compute or adjust inference. We introduce Uncertainty-Guided Inference-Time Selection, a lightweight inference time framework that disentangles aleatoric (data-driven) and epistemic (model-driven) uncertainty directly in deep feature space. Aleatoric uncertainty is estimated using a regularized global density model, while epistemic uncertainty is formed from three complementary components that capture local support deficiency, manifold spectral collapse, and cross-layer feature inconsistency. These components are empirically orthogonal and require no sampling, no ensembling, and no additional forward passes. We integrate the decomposed uncertainty into a distribution free conformal calibration procedure that yields significantly tighter prediction intervals at matched coverage. Using these components for uncertainty guided adaptive model selection reduces compute by approximately 60 percent on MOT17 with negligible accuracy loss, enabling practical self regulating visual inference. Additionally, our ablation results show that the proposed orthogonal uncertainty decomposition consistently yields higher computational savings across all MOT17 sequences, improving margins by 13.6 percentage points over the total-uncertainty baseline.
title Calibrated Decomposition of Aleatoric and Epistemic Uncertainty in Deep Features for Inference-Time Adaptation
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2511.12389