{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/53347"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/53347","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Some statistical contributions to physics-informed neural networks and energy-based models","abstract":"This dissertation presents several statistical contributions to the modeling of high-dimensional systems, focusing specifically on Physics-Informed Neural Networks (PINNs) and Energy-Based Models (EBMs). In the first part, the research examines Bayesian Physics-Informed Neural Networks (BPINNs), which integrate physical constraints directly into deep learning frameworks. This work investigates how BPINNs infer unknown parameters from noisy or incomplete observational data for PDE-based models, providing a detailed theoretical analysis of the posterior distribution’s convergence properties. Particularly, it demonstrates that, as the amount of data increases, the posterior distribution of parameters approaches a Gaussian distribution centered at the true parameters. Additionally, the study explores BPINNs in scenarios involving functional parameters that vary across space or time, enhancing understanding of their potential in modeling complex physical phenomena. In the second part, the dissertation investigates EBMs, a flexible class of statistical models that define distributions through energy functions, which are often computationally challenging due to intractable normalizing constants. This research explores short-run Markov Chain Monte Carlo (MCMC) methods by Nijkamp et al. (2019), where short, randomly initialized Markov chains approximate the gradient required for model fitting. It formalizes the theoretical connections between these short-run MCMC methods and kernel-based maximum mean discrepancy (MMD) estimations, specifically highlighting that the kernel used corresponds to the neural tangent kernel. The proposed framework is validated through practical examples, including applications to high-dimensional Gaussian graphical models and diffusion-based image generation, demonstrating improved computational efficiency and practical effectiveness compared to traditional methods.","abstract_html":"This dissertation presents several statistical contributions to the modeling of high-dimensional systems, focusing specifically on Physics-Informed Neural Networks (PINNs) and Energy-Based Models (EBMs). In the first part, the research examines Bayesian Physics-Informed Neural Networks (BPINNs), which integrate physical constraints directly into deep learning frameworks. This work investigates how BPINNs infer unknown parameters from noisy or incomplete observational data for PDE-based models, providing a detailed theoretical analysis of the posterior distribution’s convergence properties. Particularly, it demonstrates that, as the amount of data increases, the posterior distribution of parameters approaches a Gaussian distribution centered at the true parameters. Additionally, the study explores BPINNs in scenarios involving functional parameters that vary across space or time, enhancing understanding of their potential in modeling complex physical phenomena. In the second part, the dissertation investigates EBMs, a flexible class of statistical models that define distributions through energy functions, which are often computationally challenging due to intractable normalizing constants. This research explores short-run Markov Chain Monte Carlo (MCMC) methods by Nijkamp et al. (2019), where short, randomly initialized Markov chains approximate the gradient required for model fitting. It formalizes the theoretical connections between these short-run MCMC methods and kernel-based maximum mean discrepancy (MMD) estimations, specifically highlighting that the kernel used corresponds to the neural tangent kernel. The proposed framework is validated through practical examples, including applications to high-dimensional Gaussian graphical models and diffusion-based image generation, demonstrating improved computational efficiency and practical effectiveness compared to traditional methods.","abstract_has_math":false,"creators":["Sun, Yi"],"institution":"Boston University","degree_name":"Doctor of Philosophy","degree_level":"doctoral","degree_discipline":"Mathematics & Statistics","degree_department":null,"school":null,"contributors":[],"advisors":["Atchadé, Yves"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T20:13:35Z","subjects":["Statistics","Deep learning","Energy based model","Inverse problem","Machine learning","Partial differential equation","Physics-informed neural network"],"languages":["en_US"],"rights":["Attribution 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[],"person_identifier_entries":[{"key":"dc:identifier.orcid","label":"ORCID","values":["0009-0006-1022-7868"],"render_values":[{"text":"0009-0006-1022-7868","href":"https://orcid.org/0009-0006-1022-7868","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/2144/53347","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/53347","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F53347","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Atchadé, Yves"]},{"key":"dc:creator","label":"Author","values":["Sun, Yi"]},{"key":"dc:identifier.orcid","label":"ORCID","values":["0009-0006-1022-7868"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-07-17T17:39:18Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-07-17T17:39:18Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2026-07-17T16:15:38Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Mathematics & Statistics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Statistics","Deep learning","Energy based model","Inverse problem","Machine learning","Partial differential equation","Physics-informed neural network"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/53347"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["2025"]},{"key":"dc:description.abstract","label":"Abstract","values":["This dissertation presents several statistical contributions to the modeling of high-dimensional systems, focusing specifically on Physics-Informed Neural Networks (PINNs) and Energy-Based Models (EBMs). In the first part, the research examines Bayesian Physics-Informed Neural Networks (BPINNs), which integrate physical constraints directly into deep learning frameworks. This work investigates how BPINNs infer unknown parameters from noisy or incomplete observational data for PDE-based models, providing a detailed theoretical analysis of the posterior distribution’s convergence properties. Particularly, it demonstrates that, as the amount of data increases, the posterior distribution of parameters approaches a Gaussian distribution centered at the true parameters. Additionally, the study explores BPINNs in scenarios involving functional parameters that vary across space or time, enhancing understanding of their potential in modeling complex physical phenomena. In the second part, the dissertation investigates EBMs, a flexible class of statistical models that define distributions through energy functions, which are often computationally challenging due to intractable normalizing constants. This research explores short-run Markov Chain Monte Carlo (MCMC) methods by Nijkamp et al. (2019), where short, randomly initialized Markov chains approximate the gradient required for model fitting. It formalizes the theoretical connections between these short-run MCMC methods and kernel-based maximum mean discrepancy (MMD) estimations, specifically highlighting that the kernel used corresponds to the neural tangent kernel. The proposed framework is validated through practical examples, including applications to high-dimensional Gaussian graphical models and diffusion-based image generation, demonstrating improved computational efficiency and practical effectiveness compared to traditional methods."]},{"key":"dc:title","label":"Title","values":["Some statistical contributions to physics-informed neural networks and energy-based models"]}]}],"canonical_facts":{"dc:contributor.advisor":["Atchadé, Yves"],"dc:creator":["Sun, Yi"],"dc:date.accessioned":["2026-07-17T17:39:18Z"],"dc:date.available":["2026-07-17T17:39:18Z"],"dc:date.issued":["2025"],"dc:date.updated":["2026-07-17T16:15:38Z"],"dc:description":["2025"],"dc:description.abstract":["This dissertation presents several statistical contributions to the modeling of high-dimensional systems, focusing specifically on Physics-Informed Neural Networks (PINNs) and Energy-Based Models (EBMs). In the first part, the research examines Bayesian Physics-Informed Neural Networks (BPINNs), which integrate physical constraints directly into deep learning frameworks. This work investigates how BPINNs infer unknown parameters from noisy or incomplete observational data for PDE-based models, providing a detailed theoretical analysis of the posterior distribution’s convergence properties. Particularly, it demonstrates that, as the amount of data increases, the posterior distribution of parameters approaches a Gaussian distribution centered at the true parameters. Additionally, the study explores BPINNs in scenarios involving functional parameters that vary across space or time, enhancing understanding of their potential in modeling complex physical phenomena. In the second part, the dissertation investigates EBMs, a flexible class of statistical models that define distributions through energy functions, which are often computationally challenging due to intractable normalizing constants. This research explores short-run Markov Chain Monte Carlo (MCMC) methods by Nijkamp et al. (2019), where short, randomly initialized Markov chains approximate the gradient required for model fitting. It formalizes the theoretical connections between these short-run MCMC methods and kernel-based maximum mean discrepancy (MMD) estimations, specifically highlighting that the kernel used corresponds to the neural tangent kernel. The proposed framework is validated through practical examples, including applications to high-dimensional Gaussian graphical models and diffusion-based image generation, demonstrating improved computational efficiency and practical effectiveness compared to traditional methods."],"dc:identifier.orcid":["0009-0006-1022-7868"],"dc:identifier.uri":["https://hdl.handle.net/2144/53347"],"dc:language.iso":["en_US"],"dc:rights":["Attribution 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by/4.0/"],"dc:subject":["Statistics","Deep learning","Energy based model","Inverse problem","Machine learning","Partial differential equation","Physics-informed neural network"],"dc:title":["Some statistical contributions to physics-informed neural networks and energy-based models"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Mathematics & Statistics"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-18T20:13:35Z"}