Filters








827,281 Hits in 3.1 sec

Neural Importance Sampling [article]

Thomas Müller, Brian McWilliams, Fabrice Rousselle, Markus Gross, Jan Novák
2019 arXiv   pre-print
importance sampling of multi-dimensional path prefixes thereof.  ...  We propose to use deep neural networks for generating samples in Monte Carlo integration.  ...  We introduced a technique for importance sampling with neural networks.  ... 
arXiv:1808.03856v5 fatcat:zu6tbdfyuracrlmrtppq6izd2q

Neural BRDF Representation and Importance Sampling [article]

Alejandro Sztrajman, Gilles Rainer, Tobias Ritschel, Tim Weyrich
2021 arXiv   pre-print
can be mapped to parameters of an analytic BRDF for which importance sampling is known.  ...  We evaluate encoding results on isotropic and anisotropic BRDFs from multiple real-world datasets, and importance sampling performance for isotropic BRDFs mapped to two different analytic models.  ...  Other recent works, too, leveraged neural networks for importance sampling in Monte Carlo integration. Bako et al.  ... 
arXiv:2102.05963v3 fatcat:opfadpx2ujcrbii4tsy4xmpi5m

Exploring phase space with Neural Importance Sampling [article]

Enrico Bothmann, Timo Janßen, Max Knobbe, Tobias Schmale, Steffen Schumann
2020 arXiv   pre-print
We propose an importance sampling technique capable of overcoming typical deficiencies of existing approaches by incorporating neural networks.  ...  We present a novel approach for the integration of scattering cross sections and the generation of partonic event samples in high-energy physics.  ...  Using a mixture distribution for importance sampling is known as multi-channel importance sampling.  ... 
arXiv:2001.05478v3 fatcat:srhsgmcsnbem7o22cudr53u7w4

Biased Importance Sampling for Deep Neural Network Training [article]

Angelos Katharopoulos, François Fleuret
2017 arXiv   pre-print
Importance sampling has been successfully used to accelerate stochastic optimization in many convex problems.  ...  We showcase the generality of our method by testing it on both image classification and language modeling tasks using deep convolutional and recurrent neural networks.  ...  To this end, we propose a novel importance sampling scheme that accelerates the training, in theory, of any Neural Network architecture.  ... 
arXiv:1706.00043v2 fatcat:qbpckse2ufbalotlltsweyvocq

Neural BRDF Representation and Importance Sampling

Alejandro Sztrajman, Gilles Rainer, Tobias Ritschel, Tim Weyrich
2021 Computer graphics forum (Print)  
can be mapped to parameters of an analytic BRDF for which importance sampling is known.  ...  We evaluate encoding results on isotropic and anisotropic BRDFs from multiple real-world datasets, and importance sampling performance for isotropic BRDFs mapped to two different analytic models.  ...  Other recent works, too, leveraged neural networks for importance sampling in Monte Carlo integration. Bako et al.  ... 
doi:10.1111/cgf.14335 fatcat:cktjj7zsenhmni5roukp5uezwa

Importance Sampling Techniques in Neural Detector Training [chapter]

José L. Sanz-González, Diego Andina
2001 Lecture Notes in Computer Science  
In this paper, we propose and develop the use of Importance Sampling (IS) techniques in neural network training, for applications to detection in communication systems.  ...  Importance Sampling is a modified Monte Carlo technique applied to the estimation of rare event probabilities (very low probabilities).  ...  Error Probability (Pe) versus iteration number for the neural detector training, using the Importance Sampling technique.  ... 
doi:10.1007/3-540-44795-4_37 fatcat:ugbtw7axczhnxijxssuzcdfha4

Annealed Importance Sampling for Neural Mass Models

Will Penny, Biswa Sengupta, Jean Daunizeau
2016 PLoS Computational Biology  
This paper explores the use of Annealed Importance Sampling (AIS) to address these restrictions.  ...  Neural Mass Models provide a compact description of the dynamical activity of cell populations in neocortical regions.  ...  Annealed Importance Sampling Annealed Importance Sampling (AIS) [13] provides samples from a posterior density using a sequence of densities at a series of monotonically increasing inverse temperatures  ... 
doi:10.1371/journal.pcbi.1004797 pmid:26942606 pmcid:PMC4778905 fatcat:v35tyxkdvvgbxp4z3lv5i6dqua

Exploring phase space with Neural Importance Sampling

Enrico Bothmann, Timo Janßen, Max Knobbe, Tobias Schmale, Steffen Schumann
2020 SciPost Physics  
We propose an importance sampling technique capable of overcoming typical deficiencies of existing approaches by incorporating neural networks.  ...  We present a novel approach for the integration of scattering cross sections and the generation of partonic event samples in high-energy physics.  ...  Using a mixture distribution for importance sampling is known as multi-channel importance sampling.  ... 
doi:10.21468/scipostphys.8.4.069 fatcat:xtekkpkvczg75l4icopxnniy2q

Exhaustive Neural Importance Sampling applied to Monte Carlo event generation [article]

Sebastian Pina-Otey, Federico Sánchez, Thorsten Lux, Vicens Gaitan
2020 arXiv   pre-print
We present Exhaustive Neural Importance Sampling (ENIS), a method based on normalizing flows to find a suitable proposal density for rejection sampling automatically and efficiently, and discuss how this  ...  Measuring the performance of the proposal function FIG. 1 . 1 Exhaustive Neural Importance Sampling flow diagram.  ...  of functions via importance sampling [9] .  ... 
arXiv:2005.12719v1 fatcat:r65ed2evkfh6li3trlc47mfzzu

Efficient training of physics‐informed neural networks via importance sampling

Mohammad Amin Nabian, Rini Jasmine Gladstone, Hadi Meidani
2021 Computer-Aided Civil and Infrastructure Engineering  
To this end, in this paper, we study the performance of an importance sampling approach for efficient training of PINNs.  ...  Physics-Informed Neural Networks (PINNs) are a class of deep neural networks that are trained, using automatic differentiation, to compute the response of systems governed by partial differential equations  ...  The performance of implementing an importance sampling-based training has recently been evaluated on classification tasks using convolutional neural networks and recurrent neural networks [32] [33] [34  ... 
doi:10.1111/mice.12685 fatcat:mkvr6kybmvhp5gq7bewlzccctu

Adaptive Importance Sampling to Accelerate Training of a Neural Probabilistic Language Model

Y. Bengio, J.-S. Senecal
2008 IEEE Transactions on Neural Networks  
We introduce adaptive importance sampling as a way to accelerate training of the model. We show that a very significant speed-up can be obtained on standard problems.  ...  Previous work on statistical language modeling has shown that it is possible to train a feed-forward neural network to approximate probabilities over sequences of words, resulting in significant error  ...  Luckily, it can be estimated efficiently by importance sampling.  ... 
doi:10.1109/tnn.2007.912312 pmid:18390314 fatcat:h2rb64mrvnfsdkbskp3hexnfhu

Importance Sampling for Objective Function Estimations in Neural Detector Training Driven by Genetic Algorithms

Raúl Vicen-Bueno, M. Pilar Jarabo-Amores, Manuel Rosa-Zurera, José L. Sanz-González, Saturnino Maldonado-Bascón
2010 Neural Processing Letters  
Consequently, the training computational cost is unaffordable for very low objective function value estimations, and the use of Importance Sampling (IS) techniques becomes convenient.  ...  To train Neural Networks (NNs) in a supervised way, estimations of an objective function must be carried out.  ...  vectors whose pdf's are f* (x) (this pdf is known as the Importance Sampling pdf or the biasing pdf).  ... 
doi:10.1007/s11063-010-9155-8 fatcat:dgmdf7frc5g6tb6hv4azh374pq

Adaptive importance sampling for value function approximation in off-policy reinforcement learning

Hirotaka Hachiya, Takayuki Akiyama, Masashi Sugiayma, Jan Peters
2009 Neural Networks  
A common approach is to use importance sampling techniques for compensating for the bias of value function estimators caused by the difference between the data-sampling policy and the target policy.  ...  To cope with this problem, we propose using an adaptive importance sampling technique which allows us to actively control the trade-off between bias and variance.  ...  Importance Sampling Importance sampling is a general technique for dealing with the off-policy situation. Suppose we have i.i.d.  ... 
doi:10.1016/j.neunet.2009.01.002 pmid:19216050 fatcat:qmjatnuzbrhwrbsocg3fgiqwwi

Vibration Reliability Analysis of Drum Brake Using the Artificial Neural Network and Important Sampling Method

Zhou Yang, Unsong Pak, Cholu Kwon, Adil Mehmood Khan
2021 Complexity  
This research aims to evaluate the calculation accuracy and efficiency of the artificial neural network-based important sampling method (ANN-IS) on reliability of structures such as drum brakes.  ...  Because the process of FEA is time-consuming, the ANN sample size has a very important influence on the calculation efficiency.  ...  ANN sample data and testing data of the drum brake. Table 1 . FEA results of drum brake when ANN sample data size is 100. Table 2 . FEA results of drum brake when ANN sample data size is 200.  ... 
doi:10.1155/2021/5517634 fatcat:m35qw2jxgfdbnfxzkuvuj5tnh4

Neural Networks Combined with Importance Sampling Techniques for Reliability Evaluation of Explosive Initiating Device

Qi GONG, Jianguo ZHANG, Chunlin TAN, Cancan WANG
2012 Chinese Journal of Aeronautics  
function of importance sampling is constructed, and the sampling center is moved to MPP to ensure that more random sample points draw belong to the failure domain and the simulation efficiency is improved  ...  For good efficiency, based on the ideas that directional sampling reduces dimensionality and importance sampling focuses on the domain contributing to failure probability, the joint probability density  ...  MCS with Importance Sampling Based on RBF Neural Network MCS with directional importance sampling At present, the MCS method which is based on the probability theory and mathematical statistics is to  ... 
doi:10.1016/s1000-9361(11)60380-4 fatcat:nkhqyfojdjecxpktg7swuvonga
« Previous Showing results 1 — 15 out of 827,281 results