{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/129468"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/129468","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Deep models, lighter footprint compressing, explaining, and transferring Bayesian neural networks","abstract":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms","abstract_html":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms","abstract_has_math":false,"creators":["Saha, Diptarka"],"institution":"University of Illinois Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Statistics","degree_department":null,"school":null,"contributors":["Liang, Feng","Wang, Yuexi","Wang, Shulei","Chen, Yuguo"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-05-01","date_published":"2025-05-01","updated_at":"2026-07-22T22:25:05Z","subjects":["Bayesian Neural Networks","Model Compression","Feature Selection","Variational Inference","Transfer Learning"],"languages":["en","eng"],"rights":["Copyright 2025 Diptarka Saha"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/129468","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Liang, Feng","Wang, Yuexi","Wang, Shulei","Chen, Yuguo"]},{"key":"dc:creator","label":"Author","values":["Saha, Diptarka"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025-05-01","2025-05"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Statistics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Bayesian Neural Networks","Model Compression","Feature Selection","Variational Inference","Transfer Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2025 Diptarka Saha"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/129468"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms","The student, Diptarka Saha, accepted the attached license on 2025-04-29 at 12:10.","The student, Diptarka Saha, submitted this Dissertation for approval on 2025-04-29 at 12:18.","This Dissertation was approved for publication on 2025-05-01 at 06:58.","DSpace SAF Submission Ingestion Package generated from Vireo submission #22082 on 2025-10-19 at 18:19:37","Despite their widespread adoption and impressive empirical performance, modern neural networks often suffer from overparameterization, limited interpretability, and difficulty in knowledge transfer. These limitations hinder their deployment in settings where computational efficiency, transparency, and adaptability are essential. This thesis addresses these challenges through a unifying principle: building \\textit{Bayesian deep learning models with a lighter footprint}—models that retain predictive power while being more compact, interpretable, and transferable. The first contribution presents a structured compression framework for Bayesian neural networks (BNNs), where model sparsity is achieved by placing spike-and-slab priors over weights and inferring posterior inclusion probabilities. A scalable variational inference algorithm is developed that simultaneously learns network weights and their relevance, enabling principled pruning at multiple levels of granularity. The second contribution proposes a sensitivity-based feature selection method, \\texttt{SENS}, which quantifies the relevance of input features by analyzing the distribution of local sensitivities under the BNN posterior. This method provides statistically grounded feature importance scores along with confidence measures, allowing for robust interpretability and data-driven selection of relevant inputs. The third contribution introduces a semi-supervised, source-free transfer learning approach using power priors. By generating pseudo-labeled auxiliary data from a black-box model and combining it with a small labeled target set, we derive a principled posterior update that enables adaptive transfer without requiring source data or access to pretrained model internals. Each of these contributions is grounded in Bayesian methodology, equipped with theoretical guarantees, and empirically validated across a wide range of synthetic and real-world datasets. Taken together, they demonstrate that deep models need not be large, opaque, or rigid. Instead, through principled Bayesian design, we can construct neural networks that are efficient, interpretable, and resilient—qualities that are increasingly critical as machine learning systems are deployed in high-stakes, resource-constrained, or regulatory-sensitive environments."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Deep models, lighter footprint compressing, explaining, and transferring Bayesian neural networks"]}]}],"canonical_facts":{"dc:contributor":["Liang, Feng","Wang, Yuexi","Wang, Shulei","Chen, Yuguo"],"dc:creator":["Saha, Diptarka"],"dc:date":["2025-05-01","2025-05"],"dc:description":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms","The student, Diptarka Saha, accepted the attached license on 2025-04-29 at 12:10.","The student, Diptarka Saha, submitted this Dissertation for approval on 2025-04-29 at 12:18.","This Dissertation was approved for publication on 2025-05-01 at 06:58.","DSpace SAF Submission Ingestion Package generated from Vireo submission #22082 on 2025-10-19 at 18:19:37","Despite their widespread adoption and impressive empirical performance, modern neural networks often suffer from overparameterization, limited interpretability, and difficulty in knowledge transfer. These limitations hinder their deployment in settings where computational efficiency, transparency, and adaptability are essential. This thesis addresses these challenges through a unifying principle: building \\textit{Bayesian deep learning models with a lighter footprint}—models that retain predictive power while being more compact, interpretable, and transferable. The first contribution presents a structured compression framework for Bayesian neural networks (BNNs), where model sparsity is achieved by placing spike-and-slab priors over weights and inferring posterior inclusion probabilities. A scalable variational inference algorithm is developed that simultaneously learns network weights and their relevance, enabling principled pruning at multiple levels of granularity. The second contribution proposes a sensitivity-based feature selection method, \\texttt{SENS}, which quantifies the relevance of input features by analyzing the distribution of local sensitivities under the BNN posterior. This method provides statistically grounded feature importance scores along with confidence measures, allowing for robust interpretability and data-driven selection of relevant inputs. The third contribution introduces a semi-supervised, source-free transfer learning approach using power priors. By generating pseudo-labeled auxiliary data from a black-box model and combining it with a small labeled target set, we derive a principled posterior update that enables adaptive transfer without requiring source data or access to pretrained model internals. Each of these contributions is grounded in Bayesian methodology, equipped with theoretical guarantees, and empirically validated across a wide range of synthetic and real-world datasets. Taken together, they demonstrate that deep models need not be large, opaque, or rigid. Instead, through principled Bayesian design, we can construct neural networks that are efficient, interpretable, and resilient—qualities that are increasingly critical as machine learning systems are deployed in high-stakes, resource-constrained, or regulatory-sensitive environments."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/129468"],"dc:language":["en","eng"],"dc:rights":["Copyright 2025 Diptarka Saha"],"dc:subject":["Bayesian Neural Networks","Model Compression","Feature Selection","Variational Inference","Transfer Learning"],"dc:title":["Deep models, lighter footprint compressing, explaining, and transferring Bayesian neural networks"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Statistics"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:05Z"}