{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/accelerating-exact-and-approximate-inference","title":"Accelerating Exact and Approximate Inference for (Distributed) Discrete Optimization with GPUs","arxiv_id":"1608.05288","date":"2016-08-18","proceeding":null,"authors":["Ferdinando Fioretto","Enrico Pontelli","William Yeoh","Rina Dechter"],"abstract":"Discrete optimization is a central problem in artificial intelligence. The\noptimization of the aggregated cost of a network of cost functions arises in a\nvariety of problems including (W)CSP, DCOP, as well as optimization in\nstochastic variants such as the tasks of finding the most probable explanation\n(MPE) in belief networks. Inference-based algorithms are powerful techniques\nfor solving discrete optimization problems, which can be used independently or\nin combination with other techniques. However, their applicability is often\nlimited by their compute intensive nature and their space requirements. This\npaper proposes the design and implementation of a novel inference-based\ntechnique, which exploits modern massively parallel architectures, such as\nthose found in Graphical Processing Units (GPUs), to speed up the resolution of\nexact and approximated inference-based algorithms for discrete optimization.\nThe paper studies the proposed algorithm in both centralized and distributed\noptimization contexts. The paper demonstrates that the use of GPUs provides\nsignificant advantages in terms of runtime and scalability, achieving up to two\norders of magnitude in speedups and showing a considerable reduction in\nexecution time (up to 345 times faster) with respect to a sequential version.","url_abs":"http://arxiv.org/abs/1608.05288v2","url_pdf":"http://arxiv.org/pdf/1608.05288v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"accelerating-exact-and-approximate-inference","repo_url":"https://github.com/nandofioretto/GpuBE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"distributed-optimization","task_name":"Distributed Optimization"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1608.05288","atlas_url":"https://app.syntology.ai/?focus=1608.05288","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}