{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/163689"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/163689","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"GPU-accelerated Inference for Discrete Probabilistic Programs","abstract":"This thesis presents a comprehensive approach to GPU-accelerated inference for discrete probabilistic programs. We make two key contributions : (1) a factor graph IR implemented in JAX that supports variable elimination and Gibbs sampling, and (2) a modeling DSL with a compiler that lowers programs to the factor graph IR. Our system enables significant performance optimizations through static analysis of the factor graph structure. Variable elimination is optimized by reduction to tensor contraction with optimized contraction paths, while Gibbs sampling is automatically parallelized through graph coloring techniques. Empirical evaluations on standard benchmarks demonstrate orders of magnitude performance improvements over existing systems, with the parallelized Gibbs sampler showing speed-ups of up to 144x on Bayesian networks and even greater improvements for models with regular graph topologies such as Ising models and hidden Markov models.","abstract_html":"This thesis presents a comprehensive approach to GPU-accelerated inference for discrete probabilistic programs. We make two key contributions : (1) a factor graph IR implemented in JAX that supports variable elimination and Gibbs sampling, and (2) a modeling DSL with a compiler that lowers programs to the factor graph IR. Our system enables significant performance optimizations through static analysis of the factor graph structure. Variable elimination is optimized by reduction to tensor contraction with optimized contraction paths, while Gibbs sampling is automatically parallelized through graph coloring techniques. Empirical evaluations on standard benchmarks demonstrate orders of magnitude performance improvements over existing systems, with the parallelized Gibbs sampler showing speed-ups of up to 144x on Bayesian networks and even greater improvements for models with regular graph topologies such as Ising models and hidden Markov models.","abstract_has_math":false,"creators":["Ghavami, Matin"],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Mansinghka, Vikash"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-05","date_published":"2025-05","updated_at":"2026-07-22T22:21:37Z","subjects":[],"languages":[],"rights":["Attribution 4.0 International (CC BY 4.0)","Copyright retained by author(s)"],"rights_urls":["https://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/163689","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Mansinghka, Vikash"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Ghavami, Matin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-11-17T19:07:26Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-11-17T19:07:26Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Science in Electrical Engineering and Computer Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International (CC BY 4.0)","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/163689"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis presents a comprehensive approach to GPU-accelerated inference for discrete probabilistic programs. We make two key contributions : (1) a factor graph IR implemented in JAX that supports variable elimination and Gibbs sampling, and (2) a modeling DSL with a compiler that lowers programs to the factor graph IR. Our system enables significant performance optimizations through static analysis of the factor graph structure. Variable elimination is optimized by reduction to tensor contraction with optimized contraction paths, while Gibbs sampling is automatically parallelized through graph coloring techniques. Empirical evaluations on standard benchmarks demonstrate orders of magnitude performance improvements over existing systems, with the parallelized Gibbs sampler showing speed-ups of up to 144x on Bayesian networks and even greater improvements for models with regular graph topologies such as Ising models and hidden Markov models."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["S.M."]},{"key":"dc:title","label":"Title","values":["GPU-accelerated Inference for Discrete Probabilistic Programs"]}]}],"canonical_facts":{"dc:contributor.advisor":["Mansinghka, Vikash"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Ghavami, Matin"],"dc:date.accessioned":["2025-11-17T19:07:26Z"],"dc:date.available":["2025-11-17T19:07:26Z"],"dc:date.issued":["2025-05"],"dc:description.abstract":["This thesis presents a comprehensive approach to GPU-accelerated inference for discrete probabilistic programs. We make two key contributions : (1) a factor graph IR implemented in JAX that supports variable elimination and Gibbs sampling, and (2) a modeling DSL with a compiler that lowers programs to the factor graph IR. Our system enables significant performance optimizations through static analysis of the factor graph structure. Variable elimination is optimized by reduction to tensor contraction with optimized contraction paths, while Gibbs sampling is automatically parallelized through graph coloring techniques. Empirical evaluations on standard benchmarks demonstrate orders of magnitude performance improvements over existing systems, with the parallelized Gibbs sampler showing speed-ups of up to 144x on Bayesian networks and even greater improvements for models with regular graph topologies such as Ising models and hidden Markov models."],"dc:description.degree":["S.M."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/163689"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["Attribution 4.0 International (CC BY 4.0)","Copyright retained by author(s)"],"dc:rights.uri":["https://creativecommons.org/licenses/by/4.0/"],"dc:title":["GPU-accelerated Inference for Discrete Probabilistic Programs"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Science in Electrical Engineering and Computer Science"]},"updated_at":"2026-07-22T22:21:37Z"}