Distilling Calibration via Conformalized Credal Inference

Bibliographic Details
Title:	Distilling Calibration via Conformalized Credal Inference
Authors:	Huang, Jiayi, Park, Sangwoo, Paoletti, Nicola, Simeone, Osvaldo
Publication Year:	2025
Collection:	Computer Science
Subject Terms:	Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Electrical Engineering and Systems Science - Signal Processing
More Details:	Deploying artificial intelligence (AI) models on edge devices involves a delicate balance between meeting stringent complexity constraints, such as limited memory and energy resources, and ensuring reliable performance in sensitive decision-making tasks. One way to enhance reliability is through uncertainty quantification via Bayesian inference. This approach, however, typically necessitates maintaining and running multiple models in an ensemble, which may exceed the computational limits of edge devices. This paper introduces a low-complexity methodology to address this challenge by distilling calibration information from a more complex model. In an offline phase, predictive probabilities generated by a high-complexity cloud-based model are leveraged to determine a threshold based on the typical divergence between the cloud and edge models. At run time, this threshold is used to construct credal sets -- ranges of predictive probabilities that are guaranteed, with a user-selected confidence level, to include the predictions of the cloud model. The credal sets are obtained through thresholding of a divergence measure in the simplex of predictive probabilities. Experiments on visual and language tasks demonstrate that the proposed approach, termed Conformalized Distillation for Credal Inference (CD-CI), significantly improves calibration performance compared to low-complexity Bayesian methods, such as Laplace approximation, making it a practical and efficient solution for edge AI deployments. Comment: Under review
Document Type:	Working Paper
Access URL:	http://arxiv.org/abs/2501.06066
Accession Number:	edsarx.2501.06066
Database:	arXiv

FullText	Text: Availability: 0 CustomLinks: – Url: http://arxiv.org/abs/2501.06066 Name: EDS - Arxiv Category: fullText Text: View this record from Arxiv MouseOverText: View this record from Arxiv – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsarx&genre=article&issn=&ISBN=&volume=&issue=&date=20250110&spage=&pages=&title=Distilling Calibration via Conformalized Credal Inference&atitle=Distilling%20Calibration%20via%20Conformalized%20Credal%20Inference&aulast=Huang%2C%20Jiayi&id=DOI: Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries
Header	DbId: edsarx DbLabel: arXiv An: edsarx.2501.06066 RelevancyScore: 1129 AccessLevel: 3 PubType: Report PubTypeId: report PreciseRelevancyScore: 1129.22814941406
IllustrationInfo
Items	– Name: Title Label: Title Group: Ti Data: Distilling Calibration via Conformalized Credal Inference – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Huang%2C+Jiayi%22">Huang, Jiayi</searchLink><br /><searchLink fieldCode="AR" term="%22Park%2C+Sangwoo%22">Park, Sangwoo</searchLink><br /><searchLink fieldCode="AR" term="%22Paoletti%2C+Nicola%22">Paoletti, Nicola</searchLink><br /><searchLink fieldCode="AR" term="%22Simeone%2C+Osvaldo%22">Simeone, Osvaldo</searchLink> – Name: DatePubCY Label: Publication Year Group: Date Data: 2025 – Name: Subset Label: Collection Group: HoldingsInfo Data: Computer Science – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Computer+Science+-+Machine+Learning%22">Computer Science - Machine Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Science+-+Artificial+Intelligence%22">Computer Science - Artificial Intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+Engineering+and+Systems+Science+-+Signal+Processing%22">Electrical Engineering and Systems Science - Signal Processing</searchLink> – Name: Abstract Label: Description Group: Ab Data: Deploying artificial intelligence (AI) models on edge devices involves a delicate balance between meeting stringent complexity constraints, such as limited memory and energy resources, and ensuring reliable performance in sensitive decision-making tasks. One way to enhance reliability is through uncertainty quantification via Bayesian inference. This approach, however, typically necessitates maintaining and running multiple models in an ensemble, which may exceed the computational limits of edge devices. This paper introduces a low-complexity methodology to address this challenge by distilling calibration information from a more complex model. In an offline phase, predictive probabilities generated by a high-complexity cloud-based model are leveraged to determine a threshold based on the typical divergence between the cloud and edge models. At run time, this threshold is used to construct credal sets -- ranges of predictive probabilities that are guaranteed, with a user-selected confidence level, to include the predictions of the cloud model. The credal sets are obtained through thresholding of a divergence measure in the simplex of predictive probabilities. Experiments on visual and language tasks demonstrate that the proposed approach, termed Conformalized Distillation for Credal Inference (CD-CI), significantly improves calibration performance compared to low-complexity Bayesian methods, such as Laplace approximation, making it a practical and efficient solution for edge AI deployments.<br />Comment: Under review – Name: TypeDocument Label: Document Type Group: TypDoc Data: Working Paper – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2501.06066" linkWindow="_blank">http://arxiv.org/abs/2501.06066</link> – Name: AN Label: Accession Number Group: ID Data: edsarx.2501.06066
PLink	https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2501.06066
RecordInfo	BibRecord: BibEntity: Subjects: – SubjectFull: Computer Science - Machine Learning Type: general – SubjectFull: Computer Science - Artificial Intelligence Type: general – SubjectFull: Electrical Engineering and Systems Science - Signal Processing Type: general Titles: – TitleFull: Distilling Calibration via Conformalized Credal Inference Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Huang, Jiayi – PersonEntity: Name: NameFull: Park, Sangwoo – PersonEntity: Name: NameFull: Paoletti, Nicola – PersonEntity: Name: NameFull: Simeone, Osvaldo IsPartOfRelationships: – BibEntity: Dates: – D: 10 M: 01 Type: published Y: 2025
ResultId	1