Training a quantum annealing based restricted Boltzmann machine on cybersecurity data
Purdue University（普渡大学）；Oak Ridge National Laboratory（橡树岭国家实验室）；Temple University（天普大学）；Mississippi State University（密西西比州立大学）；UT-Battelle, LLC（UT-巴特尔有限责任公司）
Vivek Dixit, Raja Selvarajan, Tamer Aldwairi, Yaroslav Koshka, Mark A. Novotny, Travis S. Humble, Muhammad A. Alam, Sabre Kais
备注：
in IEEE Transactions on Emerging Topics in Computational Intelligence
发表信息：
V. Dixit et al., "Training a Quantum Annealing Based Restricted Boltzmann Machine on Cybersecurity Data," in IEEE Transactions on Emerging Topics in Computational Intelligence, doi: 10.1109/TETCI.2021.3074916
We present a real-world application that uses a quantum computer. Specifically, we train a RBM using QA for cybersecurity applications. The D-Wave 2000Q has been used to implement QA. RBMs are trained on the ISCX data, which is a benchmark dataset for cybersecurity. For comparison, RBMs are also trained using CD. CD is a commonly used method for RBM training. Our analysis of the ISCX data shows that the dataset is imbalanced. We present two different schemes to balance the training dataset before feeding it to a classifier. The first scheme is based on the undersampling of benign instances. The imbalanced training dataset is divided into five sub-datasets that are trained separately. A majority voting is then performed to get the result. Our results show the majority vote increases the classification accuracy up from 90.24% to 95.68%, in the case of CD. For the case of QA, the classification accuracy increases from 74.14% to 80.04%. In the second scheme, a RBM is used to generate synthetic data to balance the training dataset. We show that both QA and CD-trained RBM can be used to generate useful synthetic data. Balanced training data is used to evaluate several classifiers. Among the classifiers investigated, K-Nearest Neighbor (KNN) and Neural Network (NN) perform better than other classifiers. They both show an accuracy of 93%. Our results show a proof-of-concept that a QA-based RBM can be trained on a 64-bit binary dataset. The illustrative example suggests the possibility to migrate many practical classification problems to QA-based techniques. Further, we show that synthetic data generated from a RBM can be used to balance the original dataset.
Quantum annealing (QA) is a hardware-based heuristic optimization and sampling method applicable to discrete undirected graphical models. While similar to simulated annealing, QA relies on quantum, rather than thermal, effects to explore complex search spaces. For many classes of problems, QA is known to offer computational advantages over simulated annealing. Here we report on the ability of recent QA hardware to accelerate training of fully visible Boltzmann machines. We characterize the sampling distribution of QA hardware, and show that in many cases, the quantum distributions differ signi…
Learning Boltzmann Generators via Constrained Mass Transport
Autonomous Learning Robots, Karlsruhe Institute of Technology（自动化学习机器人，卡尔斯鲁厄大学技术学院）；Artificial Intelligence for Materials Sciences, Karlsruhe Institute of Technology（材料科学人工智能，卡尔斯鲁厄大学技术学院）
Multi-Mode Quantum Annealing for Generative Representation Learning with Boltzmann Priors
Department of Statistics and Data Science, Yonsei University, Seoul 03722, Republic of Korea（统计与数据科学系，延世大学，首尔03722，大韩民国）；Department of Applied Statistics, Yonsei University, Seoul 03722, Republic of Korea（应用统计学系，延世大学，首尔03722，大韩民国）；Department of Quantum Information, Yonsei University, Seoul 03722, Republic of Korea（量子信息系，延世大学，首尔03722，大韩民国）
A hybrid quantum-classical method for learning Boltzmann machines (BM) for a generative and discriminative task is presented. Boltzmann machines are undirected graphs with a network of visible and hidden nodes where the former is used as the reading site while the latter is used to manipulate visible states' probability. In Generative BM, the samples of visible data imitate the probability distribution of a given data set. In contrast, the visible sites of discriminative BM are treated as Input/Output (I/O) reading sites where the conditional probability of output state is optimized for a give…
Boltzmann machines are the basis of several deep learning methods that have been successfully applied to both supervised and unsupervised machine learning tasks. These models assume that a dataset is generated according to a Boltzmann distribution, and the goal of the training procedure is to learn the set of parameters that most closely match the input data distribution. Training such models is difficult due to the intractability of traditional sampling techniques, and proposals using quantum annealers for sampling hope to mitigate the sampling cost. However, real physical devices will inevit…
Quantum Image Denoising: A Framework via Boltzmann Machines, QUBO, and Quantum Annealing
Johns Hopkins University（约翰斯·霍普金斯大学）；NASA Ames Research Center（美国国家航空航天局艾姆斯研究中心）；Research Institute of Advanced Computer Science, USRA（大学空间研究协会高级计算机科学研究所）；NEC Corporation（日本电气股份有限公司）；National Institute of Advanced Industrial Science and Technology (AIST)（日本产业技术综合研究所）
Università degli Studi di Bologna（博洛尼亚大学）；Istituto di Fotonica e Nanotecnologie, Consiglio Nazionale delle Ricerche（国家研究委员会光子学与纳米技术研究所）；Dipartimento di Fisica, Universitá degli Studi di Milano Bicocca（米兰比可卡大学物理系）
Boltzmann Machines constitute a class of neural networks with applications to image reconstruction, pattern classification and unsupervised learning in general. Their most common variants, called Restricted Boltzmann Machines (RBMs) exhibit a good trade-off between computability on existing silicon-based hardware and generality of possible applications. Still, the diffusion of RBMs is quite limited, since their training process proves to be hard. The advent of commercial Adiabatic Quantum Computers (AQCs) raised the expectation that the implementations of RBMs on such quantum devices could inc…
(Withdrawn) This work describes an efficient user-side method of calibrating and correcting quantum annealing computers. For quantum annealing computers based on the Ising model, the method measures the residual bias of the h and J coefficients. Once measured, these biases can then be nulled in subsequent runs for any problem of interest. This method also returns a temperature for each qubit based on the measured versus the expected qubit distributions computed from a Boltzmann distribution model.
Graduate School of Information Sciences, Tohoku University（东北大学信息科学研究科）；Institute for Innovative Research, Tokyo Institute of Technology（东京工业大学创新研究所）；Sigma-i Co., Ltd.（Sigma-i 有限公司）
Quantum annealing was originally proposed as an approach for solving combinatorial optimisation problems using quantum effects. D-Wave Systems has released a production model of quantum annealing hardware. However, the inherent noise and various environmental factors in the hardware hamper the determination of optimal solutions. In addition, the freezing effect in regions with weak quantum fluctuations generates outputs approximately following a Gibbs--Boltzmann distribution at an extremely low temperature. Thus, a quantum annealer may also serve as a fast sampler for the Ising spin-glass prob…
We present an implementation of Quantum Annealing (QA) via lattice Green's function Monte Carlo (GFMC), focusing on its application to the Ising spin-glass in transverse field. In particular, we study whether or not such method is more effective than the Path-Integral Monte Carlo (PIMC) based QA, as well as classical simulated annealing (CA), previously tested on the same optimization problem. We identify the issue of importance sampling, i.e., the necessity of possessing reasonably good (variational) trial wavefunctions, as the key point of the algorithm. We have considered two possible class…
Inspired by the success of Boltzmann Machines based on classical Boltzmann distribution, we propose a new machine learning approach based on quantum Boltzmann distribution of a transverse-field Ising Hamiltonian. Due to the non-commutative nature of quantum mechanics, the training process of the Quantum Boltzmann Machine (QBM) can become nontrivial. We circumvent the problem by introducing bounds on the quantum probabilities. This allows us to train the QBM efficiently by sampling. We show examples of QBM training with and without the bound, using exact diagonalization, and compare the results…
In this paper we consider the use of certain classical analogues to quantum tunneling behavior to improve the performance of simulated annealing on a discrete spin system of the general Ising form. Specifically, we consider the use of multiple simultaneous spin flips at each annealing step as an analogue to quantum spin coherence as well as modifications of the Boltzmann acceptance probability to mimic quantum tunneling. We find that the use of multiple spin flips can indeed be advantageous under certain annealing schedules, but only for long anneal times.
Fermi National Accelerator Laboratory（费米国家加速器实验室）；Lockheed Martin Advanced Technology Center（洛克希德·马丁先进技术中心）；Kavli Institute for Cosmological Physics, University of Chicago（芝加哥大学卡弗里宇宙学物理研究所）；Department of Astronomy and Astrophysics, University of Chicago（芝加哥大学天文与天体物理学系）
We present the application of Restricted Boltzmann Machines (RBMs) to the task of astronomical image classification using a quantum annealer built by D-Wave Systems. Morphological analysis of galaxies provides critical information for studying their formation and evolution across cosmic time scales. We compress galaxy images using principal component analysis to fit a representation on the quantum hardware. Then, we train RBMs with discriminative and generative algorithms, including contrastive divergence and hybrid generative-discriminative approaches, to classify different galaxy morphologie…
The Ising model with a transverse field and an antiferromagnetic transverse interaction is represented as a matrix in the computational basis with non-zero off-diagonal elements with both positive and negative signs and thus may be regarded to be non-stoquastic. We show that the local Boltzmann factors of such a system under an appropriate Suzuki-Trotter representation can be chosen non-negative and thus may potentially be simulated classically without a sign problem if the parameter values are limited to a subspace of the whole parameter space. We then derive conditions for parameters to sati…
Boltzmann Sampling of Frustrated J1 - J2 Ising Models with Programmable Quantum Annealers
Information Systems & Modeling, Los Alamos National Laboratory（信息系统与建模，洛斯阿拉莫斯国家实验室）；Theoretical Division, Quantum & Condensed Matter Physics, Los Alamos National Laboratory（理论部，量子与凝聚态物理，洛斯阿拉莫斯国家实验室）；Center for Quantum Computing, Los Alamos National Laboratory（量子计算中心，洛斯阿拉莫斯国家实验室）
Lockheed Martin Information Systems & Global Solutions（洛克希德·马丁信息系统与全球解决方案）
In Deep Learning, a well-known approach for training a Deep Neural Network starts by training a generative Deep Belief Network model, typically using Contrastive Divergence (CD), then fine-tuning the weights using backpropagation or other discriminative techniques. However, the generative training can be time-consuming due to the slow mixing of Gibbs sampling. We investigated an alternative approach that estimates model expectations of Restricted Boltzmann Machines using samples from a D-Wave quantum annealing machine. We tested this method on a coarse-grained version of the MNIST data set. In…
Kernel Learning for Regression via Quantum Annealing Based Spectral Sampling
Graduate School of Information Sciences（信息科学研究生院）；Technology Strategy Division（技术战略部门）；Institute of Science（科学研究所）；Research and Education Institute for Semiconductors and Informatics（半导体与信息学研究教育研究所）；Sigma-i Co.（Sigma-i公司）
NASA Ames Research Center（美国国家航空航天局艾姆斯研究中心）；Bay Area Environmental Research Institute（湾区环境研究所）
We present an algorithm for learning a latent variable generative model via generative adversarial learning where the canonical uniform noise input is replaced by samples from a graphical model. This graphical model is learned by a Boltzmann machine which learns low-dimensional feature representation of data extracted by the discriminator. A quantum annealer, the D-Wave 2000Q, is used to sample from this model. This algorithm joins a growing family of algorithms that use a quantum annealing subroutine in deep learning, and provides a framework to test the advantages of quantum-assisted learnin…
By contrasting the performance of two quantum annealers operating at different temperatures, we address recent questions related to the role of temperature in these devices and their function as `Boltzmann samplers'. Using a method to reliably calculate the degeneracies of the energy levels of large-scale spin-glass instances, we are able to estimate the instance-dependent effective temperature from the output of annealing runs. Our results show that the output distributions of the annealers do not in general correspond to classical Boltzmann distributions. For the small fraction of the instan…
QB Information Technologies (1QBit)（1QB信息技术公司（1QBit））；University of British Columbia（不列颠哥伦比亚大学）；Simon Fraser University（西蒙弗雷泽大学）；University of Waterloo（滑铁卢大学）
We investigate whether quantum annealers with select chip layouts can outperform classical computers in reinforcement learning tasks. We associate a transverse field Ising spin Hamiltonian with a layout of qubits similar to that of a deep Boltzmann machine (DBM) and use simulated quantum annealing (SQA) to numerically simulate quantum sampling from this system. We design a reinforcement learning algorithm in which the set of visible nodes representing the states and actions of an optimal policy are the first and last layers of the deep network. In absence of a transverse field, our simulations…
We investigate alternative annealing schedules on the current generation of quantum annealing hardware (the D-Wave 2000Q), which includes the use of forward and reverse annealing with an intermediate pause. This work provides new insights into the inner workings of these devices (and quantum devices in general), particular into how thermal effects govern the system dynamics. We show that a pause mid-way through the anneal can cause a dramatic change in the output distribution, and we provide evidence suggesting thermalization is indeed occurring during such a pause. We demonstrate that upon pa…
Sampling from a Boltzmann distribution is NP-hard and so requires heuristic approaches. Quantum annealing is one promising candidate. The failure of annealing dynamics to equilibrate on practical time scales is a well understood limitation, but does not always prevent a heuristically useful distribution from being generated. In this paper we evaluate several methods for determining a useful operational temperature range for annealers. We show that, even where distributions deviate from the Boltzmann distribution due to ergodicity breaking, these estimates can be useful. We introduce the concep…
Mississippi State University（密西西比州立大学）；Air Force Research Laboratory（空军研究实验室）；HPC2 Center for Computational Sciences（HPC2计算科学中心）
Restricted Boltzmann Machines trained with different numbers of iterations were used to provide a diverse set of energy functions each containing many local valleys (LVs) with different energies, widths, escape barrier heights, etc. They were used to verify the previously reported possibility of using the D-Wave quantum annealer (QA) to find potentially important LVs in the energy functions of Ising spin glasses that may be missed by classical searches. For classical search, extensive simulated annealing (SA) was conducted to find as many LVs as possible regardless of the computational cost. S…
We demonstrate that the output statistics of a quantum annealing protocol run on D-Wave 2000Q can be explained by static disorder garnishing an otherwise ideal device hardware. A Boltzmann-like distribution over distinct output states emerges with increasing problem size, and significantly reduces the chances for a correct identification of the sought-after optimal solutions.
A recent Google study [Phys. Rev. X, 6:031015 (2016)] compared a D-Wave 2X quantum processing unit (QPU) to two classical Monte Carlo algorithms: simulated annealing (SA) and quantum Monte Carlo (QMC). The study showed the D-Wave 2X to be up to 100 million times faster than the classical algorithms. The Google inputs are designed to demonstrate the value of collective multiqubit tunneling, a resource available to D-Wave QPUs but not to simulated annealing. But the computational hardness in these inputs is highly localized in gadgets, with only a small amount of complexity coming from global in…
Finite-temperature criticality through quantum annealing
Max Planck Institute for the Physics of Complex Systems（马克斯·普朗克复杂系统物理研究所）；RMIT University（皇家墨尔本理工大学）；Dipartimento di Fisica e Astronomia G. Galilei, Universitá degli Studi di Padova（帕多瓦大学伽利略物理与天文学系）；Dipartimento di Ingegneria dell’Informazione, Universitá degli Studi di Padova（帕多瓦大学信息工程系）；Padua Quantum Technologies Research Center, Università degli Studi di Padova（帕多瓦大学帕多瓦量子技术研究中心）；Weizmann institute of science（魏茨曼科学研究所）
We argue that a quantum annealer at very long annealing times is likely to experience a quasistatic evolution, returning a final population that is close to a Boltzmann distribution of the Hamiltonian at a single (freeze-out) point during the annealing. Such a system is expected to correlate well with a proper quantum Monte Carlo simulation. It may also correlate with simulated annealing if at the freeze-out point the quantum energy eigenvalues are close to the classical ones. These correlations are just signatures of equilibration and do not mean that the evolution of the quantum annealer is …
Comparing Quantum Annealing and Spiking Neuromorphic Computing for Sampling Binary Sparse Coding QUBO Problems
Los Alamos National Laboratory, CCS-3 Information Sciences（洛斯阿拉莫斯国家实验室，CCS-3信息科学部门）；Harvard University, T.H. Chan School of Public Health（哈佛大学，T.H. Chan公共卫生学院）
NASA Ames Research Center（美国国家航空航天局艾姆斯研究中心）；Stinger Ghaffarian Technologies Inc.（斯汀格·加法里安技术公司）；D-Wave Systems Inc.（D-Wave系统公司）；University of Florence（佛罗伦萨大学）；Simon Fraser University（西蒙弗雷泽大学）
The development of quantum-classical hybrid (QCH) algorithms is critical to achieve state-of-the-art computational models. A QCH variational autoencoder (QVAE) was introduced in Ref. [1] by some of the authors of this paper. QVAE consists of a classical auto-encoding structure realized by traditional deep neural networks to perform inference to, and generation from, a discrete latent space. The latent generative process is formalized as thermal sampling from either a quantum or classical Boltzmann machine (QBM or BM). This setup allows quantum-assisted training of deep generative models by phy…
A successful application of quantum annealing to machine learning is training restricted Boltzmann machines (RBM). However, many neural networks for vision applications are feedforward structures, such as multilayer perceptrons (MLP). Backpropagation is currently the most effective technique to train MLPs for supervised learning. This paper aims to be forward-looking by exploring the training of MLPs using quantum annealers. We exploit an equivalence between MLPs and energy-based models (EBM), which are a variation of RBMs with a maximum conditional likelihood objective. This leads to a strate…
Restricted Boltzmann Machines (RBM) are simple statistical models defined on a bipartite graph which have been successfully used in studying more complicated many-body systems, both classical and quantum. In this work, we exploit the representation power of RBMs to provide an exact decomposition of many-body contact interactions into one-body operators coupled to discrete auxiliary fields. This construction generalizes the well known Hirsch's transform used for the Hubbard model to more complicated theories such as Pionless EFT in nuclear physics, which we analyze in detail. We also discuss po…
University of Toronto（多伦多大学）；NetraMark Corp（NetraMark公司）；Queen’s University（女王大学）
The ability to accurately classify disease subtypes is of vital importance, especially in oncology where this capability could have a life saving impact. Here we report a classification between two subtypes of non-small cell lung cancer, namely Adeno- carcinoma vs Squamous cell carcinoma. The data consists of approximately 20,000 gene expression values for each of 104 patients. The data was curated from [1] [2]. We used an amalgamation of classical and and quantum machine learning models to successfully classify these patients. We utilized feature selection methods based on univariate statisti…
University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）
Understanding how the D-Wave quantum computer could be used for machine learning problems is of growing interest. Our work evaluates the feasibility of using the D-Wave as a sampler for machine learning. We describe a hybrid system that combines a classical deep neural network autoencoder with a quantum annealing Restricted Boltzmann Machine (RBM) using the D-Wave. We evaluate our hybrid autoencoder algorithm using two datasets, the MNIST dataset and MNIST Fashion dataset. We evaluate the quality of this method by using a downstream classification method where the training is based on quantum …
