Filters








16,852 Hits in 2.1 sec

Neural Autoregressive Flows [article]

Chin-Wei Huang, David Krueger, Alexandre Lacoste, Aaron Courville
2018 arXiv   pre-print
We demonstrate that the proposed neural autoregressive flows (NAF) are universal approximators for continuous probability distributions, and their greater expressivity allows them to better capture multimodal  ...  Normalizing flows and autoregressive models have been successfully combined to produce state-of-the-art results in density estimation, via Masked Autoregressive Flows (MAF), and to accelerate state-of-the-art  ...  Related work Neural autoregressive flows are a generalization of the affine autoregressive flows introduced by Kingma et al. (2016) as inverse autoregressive flows (IAF) and further developed by Chen  ... 
arXiv:1804.00779v1 fatcat:fbnyqsf6fbda5fp5erppp7rdnm

Block Neural Autoregressive Flow [article]

Nicola De Cao, Ivan Titov, Wilker Aziz
2019 arXiv   pre-print
Recently, as an alternative to hand-crafted bijections, Huang et al. (2018) proposed neural autoregressive flow (NAF) which is a universal approximator for density functions.  ...  We propose block neural autoregressive flow (B-NAF), a much more compact universal approximator of density functions, where we model a bijection directly using a single feed-forward network.  ...  Figure 1b shows an outline of our block neural autoregressive flow.  ... 
arXiv:1904.04676v1 fatcat:ex3gsvgi6rfpbi2w4skzqboczy

Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows [article]

George Papamakarios, David C. Sterratt, Iain Murray
2019 arXiv   pre-print
SNL trains an autoregressive flow on simulated data in order to learn a model of the likelihood in the region of high posterior density.  ...  We present Sequential Neural Likelihood (SNL), a new method for Bayesian inference in simulator models, where the likelihood is intractable but simulating data from the model is possible.  ...  autoregressive flows.  ... 
arXiv:1805.07226v2 fatcat:n5vhctbtrzeg3amutvprom3r2i

HCNAF: Hyper-Conditioned Neural Autoregressive Flow and its Application for Probabilistic Occupancy Map Forecasting [article]

Geunseob Oh, Jean-Sebastien Valois
2020 arXiv   pre-print
HCNAF consists of a neural-net based conditional autoregressive flow (AF) and a hyper-network that can take large conditions in non-autoregressive fashion and outputs the network parameters of the AF.  ...  We introduce Hyper-Conditioned Neural Autoregressive Flow (HCNAF); a powerful universal distribution approximator designed to model arbitrarily complex conditional probability density functions.  ...  HCNAF We propose Hyper-Conditioned Neural Autoregressive Flow (HCNAF), a novel autoregressive flow where a transformation between X = [x 1 , x 2 , ..., x D ] ∈ R D and Z = [z 1 , z 2 , ..., z D ] ∈ R D  ... 
arXiv:1912.08111v3 fatcat:4277buemqraohf4mcivp3gzoya

Data-driven Estimation of Background Distribution through Neural Autoregressive Flows [article]

Suyong Choi, Jaehoon Lim, Hayoung Oh
2020 arXiv   pre-print
We report on a general and automatic data-driven background distribution shape estimation method using neural autoregressive flows (NAF), which is one of the deep generative learning methods.  ...  Neural Autoregressive Flows for Data-driven Shape Estimation Neural Autoregressive Flows Through deep generative methods, it is possible to approximate p( x; c) by training with data directly.  ...  In this study, we adopt neural autoregressive flows, since it is simple, but allows for universal transformation [19] .  ... 
arXiv:2008.03636v1 fatcat:pqwk2b373bgpzidr5bx3duvl5i

Gravitational-wave parameter estimation with autoregressive neural network flows [article]

Stephen R. Green, Christine Simpson, Jonathan Gair
2020 arXiv   pre-print
We introduce the use of autoregressive normalizing flows for rapid likelihood-free inference of binary black hole system parameters from gravitational-wave data with deep neural networks.  ...  We then build a more powerful latent variable model by incorporating autoregressive flows within the variational autoencoder framework.  ...  An autoregressive flow may be modeled by a neural network with masking [9] .  ... 
arXiv:2002.07656v1 fatcat:gndg4d2mxjh35b6xzfxoct344a

Improving Variational Autoencoders with Inverse Autoregressive Flow

Diederik P. Kingma, Tim Salimans, Rafal Józefowicz, Xi Chen, Ilya Sutskever, Max Welling
2016 Neural Information Processing Systems  
The proposed flow consists of a chain of invertible transformations, where each transformation is based on an autoregressive neural network.  ...  We propose a new type of normalizing flow, inverse autoregressive flow (IAF), that, in contrast to earlier published flows, scales well to high-dimensional latent spaces.  ...  We thank Harri Valpola for referring us to Gustavo Deco's relevant pioneering work on a form of inverse autoregressive flow applied to nonlinear independent component analysis.  ... 
dblp:conf/nips/KingmaSJCCSW16 fatcat:4jxc5ofgenenbjztb4ddj6fc3a

Wastewater flow forecasting model based on the nonlinear autoregressive with exogenous inputs (NARX) neural network

Khalid El Ghazouli, Jamal El Khatabi, Isam Shahrour, Aziz Soulhi
2021 H2Open Journal  
This work aimed to fill the gaps in the wastewater flow forecasting research by proposing a novel wastewater flow forecasting model (WWFFM) based on the nonlinear autoregressive with exogenous inputs neural  ...  flow forecasts.  ...  The current work aimed to fill the gaps in the wastewater flow forecasting research by proposing a novel WWFFM based on the nonlinear autoregressive with exogenous inputs neural network (NARX-NN), real-time  ... 
doi:10.2166/h2oj.2021.107 fatcat:pyxk4ewernbmhm65nsapja2lwq

HCNAF: Hyper-Conditioned Neural Autoregressive Flow and its Application for Probabilistic Occupancy Map Forecasting

Geunseob Oh, Jean-Sebastien Valois
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We introduce Hyper-Conditioned Neural Autoregressive Flow (HCNAF); a powerful universal distribution approximator designed to model arbitrarily complex conditional probability density functions.  ...  HCNAF consists of a neural-net based conditional autoregressive flow (AF) and a hyper-network that can take large conditions in nonautoregressive fashion and outputs the network parameters of the AF.  ...  HCNAF We propose Hyper-Conditioned Neural Autoregressive Flow (HCNAF), a novel autoregressive flow where a transformation between X = [x 1 , x 2 , ..., x D ] ∈ R D and Z = [z 1 , z 2 , ..., z D ] ∈ R D  ... 
doi:10.1109/cvpr42600.2020.01456 dblp:conf/cvpr/OhV20 fatcat:7eyqfncr6jgczaldaebmzzgovu

Improving Daily Peak Flow Forecasts Using Hybrid Fourier-Series Autoregressive Integrated Moving Average and Recurrent Artificial Neural Network Models

Mohammad Ebrahim Banihabib, Reihaneh Bandari, Mohammad Valipour
2020 AI  
First, a Fourier-Series Filtered Autoregressive Integrated Moving Average (FSF-ARIMA) model is applied to forecast linear behavior of daily flow time series.  ...  Second, a Recurrent Artificial Neural Network (RANN) model is utilized to forecast FSF-ARIMA model's residuals.  ...  One of the successful mathematical forecasting models are autoregressive models such as Autoregressive Integrated Moving Average (ARIMA) and multilayer perceptron Artificial Neural Network (ANN).  ... 
doi:10.3390/ai1020017 fatcat:3wznjvklebbiza64fo7yjxxjlm

Comparing neural network and autoregressive moving average techniques for the provision of continuous river flow forecasts in two contrasting catchments

Robert J. Abrahart, Linda See
2000 Hydrological Processes  
Process. 14, 2157—2172 (2000) Comparing neural network and autoregressive moving average techniques for the provision of continuous river flow forecasts in two contrasting catchments Robert J.  ...  network (NN) and autoregressive moving average (ARMA) models are compared.  ... 
doi:10.1002/1099-1085(20000815/30)14:11/12<2157::aid-hyp57>3.0.co;2-s fatcat:chmftqzq7zccrlcksenlxkfwgi

Prediction of Liquid Sodium Flow Rate through the Core of the IBR-2M Reactor Using Nonlinear Autoregressive Neural Networks

G. Ososkov, Yu. Pepelyshev, Ts. Tsogtsaikhan, Gh. Adam, J. Buša, M. Hnatič
2016 EPJ Web of Conferences  
The nonlinear autoregressive neural network (NAR) with local feedback connection has been considered as the most appropriate tool for such a prediction.  ...  This paper presents an artificial neural network method for long-term prediction of liquid sodium flow rate through the core of the IBR-2M reactor.  ...  The nonlinear autoregressive neural network (NAR) is a recurrent dynamic network based on an autoregressive model with feedback connection [7] [8] [9] .The NAR type of neural network with feedback connection  ... 
doi:10.1051/epjconf/201610802036 fatcat:2qpvniqnprdpdaxnkicvcuwwqa

An Assessment of Time Series and Autoregressive Artificial Neural Network Models, Support Vector Machine and Gene Expression Programming Models Performance in Monthly River Flow Simulation (Case Study: Kherkherechi River Basin)

Mohammad Isazadeh, Hojat Ahmadzadeh, Mohammad ALi Ghorbani, Mohammad Hassan Fazeli Fard
2018 Ì'UlÅ«m va muhandisÄ«-i Ä?byÄ?rÄ«  
بینی‬ 1-Borelli et al. 2-Feature Space 3-Vapnik and Cortes 4-Jain and Kumar 5-Kişi et al. 6-Kalteh 7-Support Vector Regression 8-Terzi and Ergin 9-Radial Basis Function Network 10-Feed-Forward Neural  ...  The other advantage of artificial neural network is the high simulation speed and the accuracy of this model in stream flow simulation.  ...  mathematical functions and support vector machine model with three kernel functions were applied for autoregressive simulation of monthly river flow.  ... 
doi:10.22055/jise.2017.13336 doaj:31a7803420ce41819432f127e4299859 fatcat:sz45xs4pwvdebo5xhmlsmiixyu

Unconstrained Monotonic Neural Networks [article]

Antoine Wehenkel, Gilles Louppe
2021 arXiv   pre-print
These transformations can be combined into powerful autoregressive flows that have been shown to be universal approximators of continuous probability distributions.  ...  We evaluate our new invertible building block within a new autoregressive flow (UMNN-MAF) and demonstrate its effectiveness on density estimation experiments.  ...  Examples include Neural Autoregressive Flows [NAF, Huang et al., 2018] and Block Neural Autoregressive Flows [B-NAF, De Cao et al., 2019] .  ... 
arXiv:1908.05164v3 fatcat:agqhtvuq2bdpbog4g7i75wtc5y

Hierarchical Autoregressive Modeling for Neural Video Compression [article]

Ruihan Yang, Yibo Yang, Joseph Marino, Stephan Mandt
2021 arXiv   pre-print
Recent work by Marino et al. (2020) showed improved performance in sequential density estimation by combining masked autoregressive flows with hierarchical latent variable models.  ...  Specifically, we view recent neural video compression methods (Lu et al., 2019; Yang et al., 2020b; Agustssonet al., 2020) as instances of a generalized stochastic temporal autoregressive transform, and  ...  Masked Autoregressive Flow (MAF).  ... 
arXiv:2010.10258v2 fatcat:76yes2d5qjfc5arzx37hpuqdbm
« Previous Showing results 1 — 15 out of 16,852 results