Estimation of effective temperatures in quantum annealers for sampling applications: A case study with possible applications in deep learning

被引:154
作者
Benedetti, Marcello [1 ,2 ,3 ]
Realpe-Gomez, John [1 ,2 ,4 ]
Biswas, Rupak [5 ]
Perdomo-Ortiz, Alejandro [1 ,6 ]
机构
[1] NASA, Ames Res Ctr, Quantum Artificial Intelligence Lab, Moffett Field, CA 94035 USA
[2] SGT Inc, 7701 Greenbelt Rd,Suite 400, Greenbelt, MD 20770 USA
[3] UCL, Dept Comp Sci, Mortimer St, London WC1E 6BT, England
[4] Univ Cartagena, Inst Matemat Aplicadas, Bolivar 130001, Colombia
[5] NASA, Ames Res Ctr, Explorat Technol Directorate, Moffett Field, CA 94035 USA
[6] Univ Calif Santa Cruz, NASA, Ames Res Ctr, Moffett Field, CA 94035 USA
关键词
INFERENCE;
D O I
10.1103/PhysRevA.94.022308
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
An increase in the efficiency of sampling from Boltzmann distributions would have a significant impact on deep learning and other machine-learning applications. Recently, quantum annealers have been proposed as a potential candidate to speed up this task, but several limitations still bar these state-of-the-art technologies from being used effectively. One of the main limitations is that, while the device may indeed sample from a Boltzmann-like distribution, quantum dynamical arguments suggest it will do so with an instance-dependent effective temperature, different from its physical temperature. Unless this unknown temperature can be unveiled, it might not be possible to effectively use a quantum annealer for Boltzmann sampling. In this work, we propose a strategy to overcome this challenge with a simple effective-temperature estimation algorithm. We provide a systematic study assessing the impact of the effective temperatures in the learning of a special class of a restricted Boltzmann machine embedded on quantum hardware, which can serve as a building block for deep-learning architectures. We also provide a comparison to k-step contrastive divergence (CD-k) with k up to 100. Although assuming a suitable fixed effective temperature also allows us to outperform one-step contrastive divergence (CD-1), only when using an instance-dependent effective temperature do we find a performance close to that of CD-100 for the case studied here.
引用
收藏
页数:13
相关论文
共 68 条
[1]   Read the fine print [J].
Aaronson, Scott .
NATURE PHYSICS, 2015, 11 (04) :291-293
[2]   Quantum adiabatic Markovian master equations [J].
Albash, Tameem ;
Boixo, Sergio ;
Lidar, Daniel A. ;
Zanardi, Paolo .
NEW JOURNAL OF PHYSICS, 2012, 14
[3]   Searching for quantum speedup in quasistatic quantum annealers [J].
Amin, Mohammad H. .
PHYSICAL REVIEW A, 2015, 92 (05)
[4]  
[Anonymous], 2015, Advances in neural information processing systems
[5]  
[Anonymous], ARXIV08110416
[6]  
[Anonymous], ARXIV160102036
[7]  
[Anonymous], ARXIV151006356
[8]  
[Anonymous], UNPUB
[9]  
[Anonymous], ARXIV150301083
[10]  
[Anonymous], ARXIV14123489