Semantic-Aware Gaussian Process Calibration with Structured Layerwise Kernels for Deep Neural Networks
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912496352231424 |
|---|---|
| author | Lee, Kyung-hwan Kim, Kyung-tae |
| author_facet | Lee, Kyung-hwan Kim, Kyung-tae |
| contents | Calibrating the confidence of neural network classifiers is essential for quantifying the reliability of their predictions during inference. However, conventional Gaussian Process (GP) calibration methods often fail to capture the internal hierarchical structure of deep neural networks, limiting both interpretability and effectiveness for assessing predictive reliability. We propose a Semantic-Aware Layer-wise Gaussian Process (SAL-GP) framework that mirrors the layered architecture of the target neural network. Instead of applying a single global GP correction, SAL-GP employs a multi-layer GP model, where each layer's feature representation is mapped to a local calibration correction. These layerwise GPs are coupled through a structured multi-layer kernel, enabling joint marginalization across all layers. This design allows SAL-GP to capture both local semantic dependencies and global calibration coherence, while consistently propagating predictive uncertainty through the network. The resulting framework enhances interpretability aligned with the network architecture and enables principled evaluation of confidence consistency and uncertainty quantification in deep models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_15987 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Semantic-Aware Gaussian Process Calibration with Structured Layerwise Kernels for Deep Neural Networks Lee, Kyung-hwan Kim, Kyung-tae Machine Learning Computer Vision and Pattern Recognition Calibrating the confidence of neural network classifiers is essential for quantifying the reliability of their predictions during inference. However, conventional Gaussian Process (GP) calibration methods often fail to capture the internal hierarchical structure of deep neural networks, limiting both interpretability and effectiveness for assessing predictive reliability. We propose a Semantic-Aware Layer-wise Gaussian Process (SAL-GP) framework that mirrors the layered architecture of the target neural network. Instead of applying a single global GP correction, SAL-GP employs a multi-layer GP model, where each layer's feature representation is mapped to a local calibration correction. These layerwise GPs are coupled through a structured multi-layer kernel, enabling joint marginalization across all layers. This design allows SAL-GP to capture both local semantic dependencies and global calibration coherence, while consistently propagating predictive uncertainty through the network. The resulting framework enhances interpretability aligned with the network architecture and enables principled evaluation of confidence consistency and uncertainty quantification in deep models. |
| title | Semantic-Aware Gaussian Process Calibration with Structured Layerwise Kernels for Deep Neural Networks |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.15987 |