336 Hits in 10.2 sec

Symmetry Breaking and Training from Incomplete Data with Radial Basis Boltzmann Machines

Marcel J. Nijman, Hilbert J. Kappen
1997 International Journal of Neural Systems  
A Radial Basis Boltzmann Machine (RBBM) is a specialized Boltzmann Machine architecture that combines feed-forward mapping with probability estimation in the input space, and for which very e cient learning  ...  We derive learning rules for the case of incomplete data, and show that they perform better on incomplete data than the traditional learning rules on a'repaired' data set. 2 Real World Computing Partnership  ...  Because of its resemblance with Radial Basis Networks, it is called a Radial Basis Boltzmann Machine (RBBM). An example of such an architecture is shown in Figure 1 .  ... 
doi:10.1142/s0129065797000318 fatcat:tpk62u64mfgwtn7gu5tj7uswe4

Physics-inspired structural representations for molecules and materials [article]

Felix Musil, Andrea Grisafi, Albert P. Bartók, Christoph Ortner, Gábor Csányi, Michele Ceriotti
2021 arXiv   pre-print
the atoms in the form of a representation that obeys the same symmetries as the properties of interest, and in general reflects the physical nature of the problem.  ...  The development of atomic-scale representations have played, and continue to play, a central role in the success of machine-learning methods that rely on this correspondence, such as interatomic potentials  ...  However, the non-trivial interaction between the radial and angular basis component makes this list incomplete.  ... 
arXiv:2101.04673v2 fatcat:hgxz3ckat5fldn54sfa54ct2cq

Machine learning with neural networks [article]

B. Mehlig
2021 arXiv   pre-print
Lecture notes for my course on machine learning with neural networks that I have given at Gothenburg University and Chalmers Technical University in Gothenburg, Sweden.  ...  Derive the contrastive divergence algorithm for training a restricted Boltzmann machine with 0/1 neurons. Bars-and-stripes data set.  ...  Find approximate decision boundaries using a radial-basis function network with m radial basis functions, for m = 5, 10, 20 and 100.  ... 
arXiv:1901.05639v3 fatcat:pyyiywuoxzds5kyc6ohqtqtd3e

Machine Learning for Predicting Thermal Transport Properties of Solids [article]

Xin Qian, Ronggui Yang
2021 arXiv   pre-print
First, machine learning is applied to solve the challenges in modeling phonon transport of crystals with defects, in amorphous materials, and at high temperatures.  ...  ., of crystalline materials with defects, of amorphous materials, and for materials at high temperatures. In the past five years, machine learning started to play a role in solving these challenges.  ...  Each tree is trained based on a subset of data, which contains randomly sampled data from the database.  ... 
arXiv:2108.12945v2 fatcat:ktoftl72dbeopk3k5mgjpi32zm

Why Unsupervised Deep Networks Generalize [article]

Anita de Mello Koch, Ellen de Mello Koch, Robert de Mello Koch
2020 arXiv   pre-print
Specializing attention mainly to autoencoders, we give an algorithm to determine the network's parameters directly from the learning data set.  ...  The resulting autoencoder almost performs as well as one trained by deep learning, and it provides an excellent initial condition for training, reducing training times by a factor between 4 and 100 for  ...  In the next section we consider unsupervised learning using a restricted Boltzmann machine (RBM).  ... 
arXiv:2012.03531v1 fatcat:yvlnrzvhazezbjhj3wv2hsloxm

Recent advances and applications of machine learning in solid-state materials science

Jonathan Schmidt, Mário R. G. Marques, Silvana Botti, Miguel A. L. Marques
2019 npj Computational Materials  
Two major questions are always the interpretability of and the physical understanding gained from machine learning models.  ...  We continue with the description of different machine learning approaches for the discovery of stable materials and the prediction of their crystal structure.  ...  From a pure prediction perspective, SVRs with radial basis function slightly outperformed Gaussian processes for training set sizes >120 materials.  ... 
doi:10.1038/s41524-019-0221-0 fatcat:egdbhhwrqjaqxbd25lafxhficq

Completed Review of Various Solar Power Forecasting Techniques Considering Different Viewpoints

Yuan-Kang Wu, Cheng-Liang Huang, Quoc-Thang Phan, Yuan-Yao Li
2022 Energies  
In comparison with the extant literature, this paper addresses more of the issues concerning the application of deep and machine learning to PV power forecasting.  ...  Based on the survey results, a complete and comprehensive solar power forecasting process must include data processing and feature extraction capabilities, a powerful deep learning structure for training  ...  However, the radial basis function graph is attenuation on both sides and radial symmetry.  ... 
doi:10.3390/en15093320 fatcat:cvvedbxhtvajzi5ibqoglb6oqy

First principles interatomic potential for tungsten based on Gaussian process regression [article]

Wojciech Jerzy Szlachta
2014 arXiv   pre-print
Our training data is based on QM information that is computed directly using density functional theory (DFT).  ...  We apply Gaussian process regression to interpolate the quantum-mechanical (QM) potential energy surface from a set of points in atomic configuration space.  ...  We then generate a GAP potential from bcc tungsten training data (with no lattice defects) and relaxation trajectories of the di-and tri-vacancies.  ... 
arXiv:1403.3291v1 fatcat:3wwg4kn4kjatto54g6nxo7ys6a

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

John A. Keith, Valentin Vassilev-Galindo, Bingqing Cheng, Stefan Chmiela, Michael Gastegger, Klaus-Robert Müller, Alexandre Tkatchenko
2021 Chemical Reviews  
We follow with a critical review of noteworthy applications that demonstrate how computational chemistry and machine learning can be used together to provide insightful (and useful) predictions in molecular  ...  Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from computational chemistry  ...  For instance, using restricted Boltzmann machines 501 or deep NNs as a basis representation of wavefunctions 105, 106, 502 in Quantum Monte Carlo calculations. −505 We have laid the general framework  ... 
doi:10.1021/acs.chemrev.1c00107 pmid:34232033 pmcid:PMC8391798 fatcat:i6mr5czvmjbf7aep3zpdt57phq

Machine Learning in Nano-Scale Biomedical Engineering [article]

Alexandros-Apostolos A. Boulogeorgos, Stylianos E. Trevlakis, Sotiris A. Tegos, Vasilis K. Papanikolaou, George K. Karagiannidis
2020 arXiv   pre-print
Machine learning (ML) empowers biomedical systems with the capability to optimize their performance through modeling of the available data extremely well, without using strong assumptions about the modeled  ...  Finally, we conclude the article with insightful discussions, that reveal research gaps and highlight possible future research directions.  ...  Moreover, they perform well even when with incomplete data sets and their models are very intuitive and easy to explain.  ... 
arXiv:2008.02195v2 fatcat:5i445iipdnag3pqukyeq2ceopy

Quantum computing enhanced machine learning for physico-chemical applications [article]

Manas Sajjan, Junxu Li, Raja Selvarajan, Shree Hari Sureshbabu, Sumit Suresh Kale, Rishabh Gupta, Sabre Kais
2021 arXiv   pre-print
Machine learning (ML) has emerged into formidable force for identifying hidden but pertinent patterns within a given data set with the objective of subsequent generation of automated predictive behavior  ...  from ML and in turn can potentially accelerate the growth of such algorithms.  ...  The data modeling and training of Boltzmann machine is carried out using the equilibrium thermal states of the transverse Ising type Hamiltonian.However, the training procedure used in 232 suffered from  ... 
arXiv:2111.00851v1 fatcat:i2caiglszvbufbyfmf3cwkcduu

Multi-Fault Diagnosis Of Industrial Rotating Machines Using Data-Driven Approach: A Review Of Two Decades Of Research [article]

Shreyas Gawde, Shruti Patil, Satish Kumar, Pooja Kamat, Ketan Kotecha, Ajith Abraham
2022 arXiv   pre-print
There is a need to systematically cover all the aspects right from sensor selection, data acquisition, feature extraction, multi-sensor data fusion to the systematic review of AI techniques employed in  ...  With Artificial Intelligence (AI) advancement, data-driven approach for predictive maintenance is taking a new flight towards smart manufacturing.  ...  Deep Boltzmann Machines (DBM) [88] .  ... 
arXiv:2206.14153v1 fatcat:qqcznet7cjb63ioed3rh77cv5q

Gaussian Process Regression for Materials and Molecules

Volker L. Deringer, Albert P. Bartók, Noam Bernstein, David M. Wilkins, Michele Ceriotti, Gábor Csányi
2021 Chemical Reviews  
Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed.  ...  We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry.  ...  indices, a and a′), with the square of the radial basis expansion limit (due to the two indices n and n′), and linearly with the angular basis expansion limit (due to the index l).  ... 
doi:10.1021/acs.chemrev.1c00022 pmid:34398616 pmcid:PMC8391963 fatcat:ns54wrx4nzcw3lsztdkxblpeeq

Neural-Symbolic Learning and Reasoning: A Survey and Interpretation [article]

Tarek R. Besold, Artur d'Avila Garcez, Sebastian Bader, Howard Bowman, Pedro Domingos, Pascal Hitzler, Kai-Uwe Kuehnberger, Luis C. Lamb, Daniel Lowd, Priscila Machado Vieira Lima, Leo de Penning, Gadi Pinkas, Hoifung Poon (+1 others)
2017 arXiv   pre-print
Such systems have shown promise in a range of applications, including computational biology, fault diagnosis, training and assessment in simulators, and software verification.  ...  In addition, efforts in computer science research have led to the development of cognitive computational systems integrating machine learning and automated reasoning.  ...  In the corresponding simulations, high-order variants of Boltzmann Machines are therefore used, as they are faster to simulate and have a smaller search space than the "standard" Boltzmann machines with  ... 
arXiv:1711.03902v1 fatcat:3fod6z4oevhplpv2hzlkguyqiu

2022 Review of Data-Driven Plasma Science [article]

Rushil Anirudh, Rick Archibald, M. Salman Asif, Markus M. Becker, Sadruddin Benkadda, Peer-Timo Bremer, Rick H.S. Budé, C.S. Chang, Lei Chen, R. M. Churchill, Jonathan Citrin, Jim A Gaffney (+51 others)
2022 arXiv   pre-print
Therefore, it is imperative to train machines to analyze and interpret (eventually) such data as intelligently as humans but far more efficiently in quantity.  ...  A large amount of data and machine learning algorithms go hand in hand. Most plasma data, whether experimental, observational or computational, are generated or collected by machines today.  ...  In reality, more than two environmental parameters FIG. 36 : Left side from top to bottom: An atom i and its surrounding atoms, one of the corresponding radial vectors and one radial symmetry function  ... 
arXiv:2205.15832v1 fatcat:fxsl6gl3fncnhpoj76defxoc3a
« Previous Showing results 1 — 15 out of 336 results