FPGA Acceleration of Probabilistic Sentential Decision Diagrams with High-level Synthesis
-
Published:2023-03-11
Issue:2
Volume:16
Page:1-22
-
ISSN:1936-7406
-
Container-title:ACM Transactions on Reconfigurable Technology and Systems
-
language:en
-
Short-container-title:ACM Trans. Reconfigurable Technol. Syst.
Author:
Choi Young-Kyu1ORCID,
Santillana Carlos2ORCID,
Shen Yujia2ORCID,
Darwiche Adnan2ORCID,
Cong Jason2ORCID
Affiliation:
1. Inha University, South Korea and University of California, Los Angeles, California
2. University of California, Los Angeles, California
Abstract
Probabilistic Sentential Decision Diagrams (PSDDs) provide efficient methods for modeling and reasoning with probability distributions in the presence of massive logical constraints. PSDDs can also be synthesized from graphical models such as Bayesian networks (BNs) therefore offering a new set of tools for performing inference on these models (in time linear in the PSDD size). Despite these favorable characteristics of PSDDs, we have found multiple challenges in PSDD’s FPGA acceleration. Problems include limited parallelism, data dependency, and small pipeline iterations. In this article, we propose several optimization techniques to solve these issues with novel pipeline scheduling and parallelization schemes. We designed the PSDD kernel with a high-level synthesis (HLS) tool for ease of implementation and verified it on the Xilinx Alveo U250 board. Experimental results show that our methods improve the baseline FPGA HLS implementation performance by 2,200X and the multicore CPU implementation by 20X. The proposed design also outperforms state-of-the-art BN and Sum Product Network (SPN) accelerators that store the graph information in memory.
Funder
Inha University Research Grant, National Research Foundation (NRF) Grant funded by Korea Ministry of Science and ICT
US NSF Grant on RTML: Large: Acceleration to Graph-Based Machine Learning
Xilinx Heterogeneous Accelerated Compute Cluster (HACC) Program
Publisher
Association for Computing Machinery (ACM)
Subject
General Computer Science
Reference45 articles.
1. Berkeley. 2008. BEE3 (Berkeley Emulation Engine). Retrieved 24 July, 2022 from https://www.microsoft.com/en-us/research/project/bee3/.
2. Simone Bova. 2016. SDDs are exponentially more succinct than OBDDs. In Proceedings of the AAAI Conference on Artificial Intelligence. AAAI Press, 929–935.
3. VIBNN
4. Hei Chan and Adnan Darwiche. 2006. On the robustness of most probable explanations. In Proceedings of the 22nd Conference in Uncertainty in Artificial Intelligence.
5. On probabilistic inference by weighted model counting
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献