25,624 Hits in 5.3 sec

Efficient learning in sparsely connected Boltzmann machines [chapter]

Marcel J. Nijman, Hilbert J. Kappen
1996 Lecture Notes in Computer Science  
The resulting speed-up is of immediate relevance for the speed-up of Boltzmann Machine learning rules, especially for networks with a sparse connectivity.  ...  We present a heuristical procedure for efficient estimation of the partition function in the Boltzmann distribution.  ...  Conclusions We have shown how the partition function of a Boltzmann Machine with a sparsely connected structure can be computed efficiently, which leads to efficient learning rules.  ... 
doi:10.1007/3-540-61510-5_11 fatcat:rrpsi2apqnctvd6xskjvbfvsfm

Neural network representation of tensor network and chiral states [article]

Yichen Huang, Joel E. Moore
2021 arXiv   pre-print
These results demonstrate the power of Boltzmann machines.  ...  We study the representational power of Boltzmann machines (a type of neural network) in quantum many-body systems.  ...  Very recently, we became aware of some related papers [28] [29] [30] , which studied the relationship between neural and tensor network states using different methods.  ... 
arXiv:1701.06246v2 fatcat:mewtck6efjdsfmwwdzbd65n5f4

Boltzmann Machine and Hyperbolic Activation Function in Higher Order Network

Saratha Sathasivam, Muraly Velavan
2014 Modern Applied Science  
However, by using Boltzmann Machines and hyperbolic tangent activation function this problem can be overcome.  ...  A Boltzmann machine, like a Hopfield network, is a network of units with an "energy" defined for the network. Unlike Hopfield networks, binary units of Boltzmann Machine are stochastic.  ...  Thus, Boltzmann Machines had introduced to overcome this problem. A Boltzmann machine is a network of units which are fully interconnected by bidirectional connections with symmetric weights.  ... 
doi:10.5539/mas.v8n3p140 fatcat:2mtcsjrvsfdpnnbd5o6vpmwbvi

Neural-Network Quantum States: A Systematic Review [article]

David R. Vivas, Javier Madroñero, Victor Bucheli, Luis O. Gómez, John H. Reina
2022 arXiv   pre-print
In the context of quantum many-body physics, its intersection with machine learning has configured a high-impact interdisciplinary field of study; with the arise of recent seminal contributions that have  ...  Here, a systematic review of literature regarding Neural-Network Quantum States is presented.  ...  In 17 , the authors review and discuss the impact of Restricted-Boltzmann machines in quantum physics, with special emphasis on NQS and condensed matter.  ... 
arXiv:2204.12966v1 fatcat:p6m3h5fqwnfara5fskhgyzmama

Deep Boltzmann Machines and the Centering Trick [chapter]

Grégoire Montavon, Klaus-Robert Müller
2012 Lecture Notes in Computer Science  
Deep Boltzmann machines are in principle powerful models for extracting the hierarchical structure of data.  ...  We test the algorithm on real data and demonstrate that our suggestion, the centered deep Boltzmann machine, learns a hierarchy of increasingly abstract representations and a better generative model of  ...  Example of Boltzmann machines with different structures are shown in Figure 1 .  ... 
doi:10.1007/978-3-642-35289-8_33 fatcat:i3mxvfjivrhm3budikyn6vpwce

Interpretable Sparse High-Order Boltzmann Machines

Martin Renqiang Min, Xia Ning, Chao Cheng, Mark Gerstein
2014 International Conference on Artificial Intelligence and Statistics  
Fully-observable high-order Boltzmann Machines are capable of identifying explicit highorder feature interactions theoretically.  ...  To the best of our knowledge, SHBM is the first working Boltzmann Machine with explicit highorder feature interactions applied to realworld problems.  ...  Section 7 concludes the paper with some discussions and future work. 2 Sparse High-order Boltzmann Machines Review of Boltzmann Machines and High-order Boltzmann Machines In this section, we review traditional  ... 
dblp:conf/aistats/MinNCG14 fatcat:krx3remjybanjmgfwglichp74y

Learning thermodynamics with Boltzmann machines

Giacomo Torlai, Roger G. Melko
2016 Physical review B  
The trained Boltzmann machine is then used to generate spin states, for which we compare thermodynamic observables to those computed by direct MC sampling.  ...  Through unsupervised learning, we train the Boltzmann machine on data sets constructed with spin configurations importance-sampled from the partition function of an Ising Hamiltonian at different temperatures  ...  A natural candidate neural network for this task is a Boltzmann machine. A Boltzmann machine is a stochastic neural network, composed of neuron-like nodes forming a network with undirected edges.  ... 
doi:10.1103/physrevb.94.165134 fatcat:rb2ylo3drjcpzhvrvpkf6sff4e

An Overview of Hopfield Network and Boltzmann Machine

Saratha Sathasivam, Abdu Masanawa Sagir
2014 International journal of computational and electronics aspects in engineering  
The two well-known and commonly used types of recurrent neural networks, Hopfield neural network and Boltzmann machine have different structures and characteristics.  ...  This study gives an overview of Hopfield network and Boltzmann machine in terms of architectures, learning algorithms, comparison between these two networks from several different aspects as well as their  ...  A Boltzmann machine, like a Hopfield network, is a network of units with an "energy" defined for the network.  ... 
doi:10.26706/ijceae.1.1.20141205 fatcat:ost6kvnxzjcqrdi4qa6oww3ybi

Deep Restricted Boltzmann Networks [article]

Hengyuan Hu and Lisheng Gao and Quanbin Ma
2016 arXiv   pre-print
Restricted Boltzmann machine (RBM) is one of such models that is simple but powerful.  ...  The most famous ones among them are deep belief network, which stacks multiple layer-wise pretrained RBMs to form a hybrid model, and deep Boltzmann machine, which allows connections between hidden units  ...  a deep Boltzmann machine.  ... 
arXiv:1611.07917v1 fatcat:ovfy5n7lzbfw3mpesop5ulgvym

Connectionist learning of belief networks

Radford M. Neal
1992 Artificial Intelligence  
Experimental results show that, as a result, learning in a sigmoid belief network can be faster than in a Boltzmann machine.  ...  Due to the directed nature of the connections in a belief network, however, the "negative phase" of Boltzmann machine learning is unnecessary.  ...  This paper begins with reviews of Boltzmann machines and belief networks.  ... 
doi:10.1016/0004-3702(92)90065-6 fatcat:qkr3vdvj7zgoblkqflw2mzwyhq

Linear Neurons and Their Learning Algorithms

Ying Liu
2018 Journal of Computer Science and Information Technology  
First, we briefly review the Boltzmann Machine and the fact that the invariant distributions of the Boltzmann Machine generate Markov chains.  ...  The building block of a DBN is a probabilistic model called a Restricted Boltzmann machine (RBM), used to represent one layer of the model.  ...  Each configuration of a Boltzmann machine is a state of the Markov chain. A Boltzmann machine has a stable distribution. Let T be the parameter space of a family of Boltzmann machines.  ... 
doi:10.15640/jcsit.v6n2a1 fatcat:gdi427au6fhoxhwglztzo7jpxa

Boltzmann Machine and Hyperbolic activation function in Higher Order Neuro Symbolic Integration

Muraly Velavan, Zainor Ridzuan bin Yahya, Mohamad Nazri bin Abdul Halif, Saratha Sathasivam
2015 International journal of computational and electronics aspects in engineering  
Index terms -Boltzmann machine, agent based modelling and hyperbolic tangent activation function I.  ...  Hereby we employed Boltzmann Machines and hyperbolic tangent activation function to increased the performance of neuro symbolic integration. We used agent based modelling to model this problem.  ...  A Boltzmann machine is made up of stochastic units.  ... 
doi:10.26706/ijceae.1.1.20150103 fatcat:6piw6wvbjbapjji2g7rpu4yc7a

Securing 5G-IoT Device Connectivity and Coverage Using Boltzmann Machine Keys Generation

Anand Singh Rajawat, Pradeep Bedi, S. B. Goyal, Piyush Kumar Shukla, Sajjad Shaukat Jamal, Adel R. Alharbi, Amer Aljaedi, Vijay Kumar
2021 Mathematical Problems in Engineering  
In this research, we proposed a Boltzmann machine (BMKG)-based encryption algorithm for securing 5G-enabled IoT device network environment and compared various asymmetric algorithms for key exchange.  ...  Because purposeful faults can quickly render the entire network dysfunctional, they are more difficult to identify than unexpected failures.  ...  for generating other machine states.  ... 
doi:10.1155/2021/2330049 fatcat:zhebkqqixfapzlmzi3gkupyxju

On the challenges of using D-Wave computers to sample Boltzmann Random Variables [article]

Thomas Pochart, Paulin Jacquot, Joseph Mikael
2021 arXiv   pre-print
Sampling random variables following a Boltzmann distribution is an NP-hard problem involved in various applications such as training of Boltzmann machines, a specific kind of neural network.  ...  We detail the various obstacles and explain the remaining difficulties in solving the sampling problem on a D-wave machine.  ...  INTRODUCTION Sampling Boltzmann-distributed variables associated with an Ising model is an NP-hard problem [1] , which is notably involved in the training of the so-called Boltzmann machines (BM), a particular  ... 
arXiv:2111.15295v2 fatcat:ey2uzsnrwnbm3gtccht5yjvktq

Boltzmann Sampling by Degenerate Optical Parametric Oscillator Network for Structure-Based Virtual Screening

Hiromasa Sakaguchi, Koji Ogata, Tetsu Isomura, Shoko Utsunomiya, Yoshihisa Yamamoto, Kazuyuki Aihara
2016 Entropy  
Results of numerical simulation demonstrate that the DOPO network efficiently identifies a set of appropriate ligand molecules according to the Boltzmann sampling law.  ...  This procedure takes a known structure of a protein-ligand complex as input, and structurally similar compounds with the query ligand are designed in consideration with all possible combinations of atomic  ...  All degenerate states are found with almost equal probability at C/A = 0.5.  ... 
doi:10.3390/e18100365 fatcat:wdp7w4pl5vabvbgdqgqgazyrua
« Previous Showing results 1 — 15 out of 25,624 results