Boltzmann machines as two-dimensional tensor networks
文献类型:期刊论文
作者 | Li, Sujie1; Pan, Feng1; Zhou, Pengfei1; Zhang, Pan2,3![]() |
刊名 | PHYSICAL REVIEW B
![]() |
出版日期 | 2021 |
卷号 | 104期号:7页码:75154 |
ISSN号 | 2469-9950 |
DOI | 10.1103/PhysRevB.104.075154 |
英文摘要 | Restricted Boltzmann machines (RBMs) and deep Boltzmann machines (DBMs) are important models in machine learning, and recently found numerous applications in quantum many-body physics. We show that there are fundamental connections between them and tensor networks. In particular, we demonstrate that any RBM and DBM can be exactly represented as a two-dimensional tensor network. This representation gives characterizations of the expressive power of RBMs and DBMs using entanglement structures of the tensor networks, and also provides an efficient tensor network contraction algorithm for the computing partition function of RBMs and DBMs. Using numerical experiments, we show that the proposed algorithm is more accurate than the state-of-the-art machine learning methods in estimating the partition function of RBMs and DBMs, and have potential applications in training DBMs for general machine learning tasks. |
学科主题 | Materials Science ; Physics |
语种 | 英语 |
源URL | [http://ir.itp.ac.cn/handle/311006/27278] ![]() |
专题 | 理论物理研究所_理论物理所1978-2010年知识产出 |
作者单位 | 1.Chinese Acad Sci, Inst Theoret Phys, CAS Key Lab Theoret Phys, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Phys Sci, Beijing 100049, Peoples R China 3.UCAS, Hangzhou Inst Adv Study, Sch Fundamental Phys & Math Sci, Hangzhou 310024, Peoples R China 4.Int Ctr Theoret Phys Asia Pacific, Beijing, Peoples R China |
推荐引用方式 GB/T 7714 | Li, Sujie,Pan, Feng,Zhou, Pengfei,et al. Boltzmann machines as two-dimensional tensor networks[J]. PHYSICAL REVIEW B,2021,104(7):75154. |
APA | Li, Sujie,Pan, Feng,Zhou, Pengfei,&Zhang, Pan.(2021).Boltzmann machines as two-dimensional tensor networks.PHYSICAL REVIEW B,104(7),75154. |
MLA | Li, Sujie,et al."Boltzmann machines as two-dimensional tensor networks".PHYSICAL REVIEW B 104.7(2021):75154. |
入库方式: OAI收割
来源:理论物理研究所
浏览0
下载0
收藏0
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。