{"type":"ScholarlyArticle","title":"Deep Bucket Elimination","authors":[{"name":"Yasaman Razeghi","slug":"yasaman-razeghi","given":"Yasaman","family":"Razeghi","affiliation":""},{"name":"Kalev Kask","slug":"kalev-kask","given":"Kalev","family":"Kask","affiliation":""},{"name":"Yadong Lu","slug":"yadong-lu","given":"Yadong","family":"Lu","affiliation":""},{"name":"Pierre Baldi","slug":"pierre-baldi","given":"Pierre","family":"Baldi","affiliation":""},{"name":"Sakshi Agarwal","slug":"sakshi-agarwal","given":"Sakshi","family":"Agarwal","affiliation":""},{"name":"Rina Dechter","slug":"rina-dechter","given":"Rina","family":"Dechter","affiliation":""}],"abstract":"Bucket Elimination (BE) is a universal inference scheme that can solve most tasks over probabilistic and deterministic graphical models exactly.\r\nHowever, it often requires exponentially high levels of memory (in the induced-width) preventing its execution. In the spirit of exploiting Deep Learning for inference tasks, in this paper, we will use neural networks to approximate BE. \r\nThe resulting Deep Bucket Elimination (DBE) algorithm is developed for computing the partition function.\r\nWe provide a proof-of-concept empirically using instances from several different benchmarks, showing that DBE can be a more accurate approximation than current state-of-the-art approaches for approximating  BE (e.g. the mini-bucket schemes), especially when problems are sufficiently hard.","doi":"10.24963/ijcai.2021/582","url":"https://ijcai.org/proceedings/2021/582","pdf":"https://ijcai.org/proceedings/2021/0582.pdf","markdown":"https://ijcai.org/proceedings/2021/582.md","bibtex":"https://ijcai.org/proceedings/2021/bibtex/582","volume":{"slug":"2021","title":"Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2021","url":"https://ijcai.org/proceedings/2021/","year":2021,"location":"Montreal","dates":"19-27 August 2021","editors":"Prof. Zhi-Hua Zhou","isbn":"978-0-9992411-9-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2021"},"section":"Uncertainty in AI","track":"Main Track","number":582,"pages":{"first":4235,"last":4242},"keywords":["Uncertainty in AI: Approximate Probabilistic Inference","Uncertainty in AI: Exact Probabilistic Inference"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}