1,449 Hits in 6.3 sec

Nonparametric Conditional Density Estimation In A Deep Learning Framework For Short-Term Forecasting [article]

David B. Huberman, Brian J. Reich, Howard D. Bondell
2020 arXiv   pre-print
Short-term forecasting is an important tool in understanding environmental processes.  ...  In this paper, we incorporate machine learning algorithms into a conditional distribution estimator for the purposes of forecasting tropical cyclone intensity.  ...  Bayesian nonparametric mixture modeling is another common conditional density estimation approach.  ... 
arXiv:2008.07653v1 fatcat:kk3bh6bvjbgvrkpi5ccv6wmmki

Probabilistic and Deterministic Wind Speed Prediction: Ensemble Statistical Deep Regression Network

Solmaz Farahbod, Taher Niknam, Mohammad Mohammadi, Jamshid Aghaei, Sattar Shojaeiyan
2022 IEEE Access  
Hanna Niemelä for her assistance with proofreading of the paper and comments that greatly improved the manuscript.  ...  These issues motivate us to establish a probabilistic approach for wind speed forecasting based on a combination of advanced deep networks and the nonparametric probability density function (PDF) estimators  ...  In the nonparametric PDF estimators, kernel density estimation (KDE) is well known as an easy implementation of nonparametric PDF estimation, and it can estimate the PDFs without any prior knowledge of  ... 
doi:10.1109/access.2022.3171610 fatcat:c5votlxstrcszmq3vyjtswbyiy

A deep attention-driven model to forecast solar irradiance

Abdelkader Dairi, Fouzi Harrou, Ying Sun
2021 2021 IEEE 19th International Conference on Industrial Informatics (INDIN)  
This paper introduces an innovative deep attentiondriven model for solar irradiance forecasting.  ...  Results confirm the superior performance of the proposed model for solar irradiance forecasting over the other models (i.e., RNN, GRU, LSTM, and BiLSTM).  ...  In [12] , a nonparametric estimator for estimating solar irradiance probability density has been introduced based on local linear regression and a root transformation approach.  ... 
doi:10.1109/indin45523.2021.9557405 fatcat:plucedbnf5a47gc5j6ug7bmxpy

Traffic Flow Prediction using Machine Learning Techniques - A Systematic Literature Review

Sigma Sathyan, N. Jagadeesha S.
2022 Zenodo  
The information acquired will be useful to expand on existing theories and frameworks or to develop a new technique or modify to improve the accuracy of TFP.  ...  The collected information is then reviewed to discover possible key areas of concern in the TFP and ITS. Findings/Results: Traffic management in cities is important for smooth traffic flow.  ...  For this, a unique deep learning framework called Traffic Graph Convolutional Long Short-Term Memory Neural Network (TGC-LSTM) that learns the interactions between highways in the traffic graph and forecasts  ... 
doi:10.5281/zenodo.6479157 fatcat:mei5u4hf5bft5mpgaawjfv77ru

Smooth Pinball Neural Network for Probabilistic Forecasting of Wind Power [article]

Kostas Hatalis, Alberto J. Lamadrid, Katya Scheinberg, Shalinee Kishore
2017 arXiv   pre-print
Whereas point forecasting provides a single expected value, probabilistic forecasts provide more information in the form of quantiles, prediction intervals, or full predictive densities.  ...  This paper analyzes the effectiveness of a novel approach for nonparametric probabilistic forecasting of wind power that combines a smooth approximation of the pinball loss function with a neural network  ...  Forecasting horizons can be categorized into several scales: very short term (several seconds or minutes ahead), short term (several hours to days ahead), long term (weeks or months ahead), and seasonal  ... 
arXiv:1710.01720v1 fatcat:lnvxestu6vc4zi6fwix36p7oom

A review of short‐term wind power probabilistic forecasting and a taxonomy focused on input data

Ioannis K. Bazionis, Panagiotis A. Karafotis, Pavlos S. Georgilakis
2021 IET Renewable Power Generation  
A review of state-of-the-art short-term wind power probabilistic forecasting models is the focus here.  ...  Future directions in the field of short-term wind power probabilistic forecasting are also proposed.  ...  2017 A data-driven multi-model methodology along with a deep-feature selection is proposed for short-term wind forecasting.  ... 
doi:10.1049/rpg2.12330 fatcat:wk2piydl4fcrvd22jq5qsc7aem

Short-Term Density Forecasting of Low-Voltage Load using Bernstein-Polynomial Normalizing Flows [article]

Marcel Arpogaus, Marcus Voss, Beate Sick, Mark Nigge-Uricher, Oliver Dürr
2022 arXiv   pre-print
We propose an approach for flexible conditional density forecasting of short-term load based on Bernstein polynomial normalizing flows, where a neural network controls the parameters of the flow.  ...  However, high fluctuations and increasing electrification cause huge forecast variability, not reflected in traditional point estimates.  ...  In Section II we give a short introduction on NFs and describe our Bernstein-Polynomial Normalizing Flow (BNF) approach for short-term density forecasting of LV loads.  ... 
arXiv:2204.13939v1 fatcat:jf6s3bnxfvbm7aq36gxwqbv3gq

A multi-task deep learning model for short-term taxi demand forecasting considering spatiotemporal dependences

Huimin Luo, Jianming Cai, Kunpeng Zhang, Ruihang Xie, Liang Zheng
2020 Journal of Traffic and Transportation Engineering (English ed. Online)  
A multi-task deep learning model for short-term taxi demand forecasting considering spatiotemporal dependences, Journal of Traffic and Transportation Engineering (English Edition), https://doi.9 Highlights  ...  With the 18 consideration of spatiotemporal dependences, this study proposes a multi-task deep learning (MTDL) model 19 to predict short-term taxi demand in multi-zone level.  ...  This paper 27 explores the short-term taxi demand forecasting via a novel multi-task deep learning model considering machine learning algorithms.  ... 
doi:10.1016/j.jtte.2019.07.002 fatcat:2vgiq3zwhjelxm2bkel6dzwtz4

A Deep Generative Adversarial Architecture for Network-Wide Spatial-Temporal Traffic State Estimation [article]

Yunyi Liang, Zhiyong Cui, Yu Tian, Huimiao Chen, Yinhai Wang
2018 arXiv   pre-print
This study proposes a deep generative adversarial architecture (GAA) for network-wide spatial-temporal traffic state estimation.  ...  It consists of two Long Short-Term Memory Neural Networks (LSTM NNs) which capture correlation in time and space among traffic flow and traffic density.  ...  Deep learning for short-term traffic flow prediction. Transportation Research Part C: Emerging Technologies, 2017, 79: 1-17. 15. Ma X, Tao Z, Wang Y, et al.  ... 
arXiv:1801.03818v1 fatcat:ft5zxclcdnfqphkzfgquzs4ibi

Sparse Gaussian Process Regression for Landslide Displacement Time-Series Forecasting

Weiqi Yang, Yuran Feng, Jian Wan, Lingling Wang
2022 Frontiers in Earth Science  
The experimental results confirmed the superiority of the sparse Gaussian process in the modeling of landslide displacement series in terms of forecasting accuracy, uncertainty quantification, and robustness  ...  In this study, a probabilistic landslide displacement forecasting model based on the quantification of epistemic uncertainty is proposed.  ...  KDE is a classic nonparametric estimation method used in a wide variety of probabilistic forecasting tasks (Botev et al., 2010; Kim & Scott 2012) .  ... 
doi:10.3389/feart.2022.944301 fatcat:am4lalyn6bb7naheistc5mkcsm

Framework of 2D KDE and LSTM-Based Forecasting for Cost-Effective Inventory Management in Smart Manufacturing

Myungsoo Kim, Jaehyeong Lee, Chaegyu Lee, Jongpil Jeong
2022 Applied Sciences  
The framework makes the prediction using the point prediction model by means of LSTM(Long Short Term Memory), the 2D kernel density function, and the prediction result reflecting inventory-management cost  ...  For this reason, studies have been conducted to design predictive models using machine learning in various industries.  ...  In this paper, we first design a demand-prediction model based on LSTM (Long-Short-Term Memory).  ... 
doi:10.3390/app12052380 fatcat:gbmfe5325jec3nq2d6guddzz3m

Traffic Information Enrichment: Creating Long-Term Traffic Speed Prediction Ensemble Model for Better Navigation through Waypoints

Milan Simunek, Zdenek Smutny
2020 Applied Sciences  
Traffic speed prediction for a selected road segment from a short-term and long-term perspective is among the fundamental issues of intelligent transportation systems (ITS).  ...  However, no satisfactory solution has been found for the issue of a long-term prediction for days and weeks using the vast spatial and temporal data.  ...  A typical solution in this area uses deep belief networks [39] , long short-term memory neural networks [40] , and their combination [41] .  ... 
doi:10.3390/app11010315 fatcat:ajzndas47ndktebjp5wyrqvsdu

Weather-Aware Long-Range Traffic Forecast Using Multi-Module Deep Neural Network

Seungyo Ryu, Dongseung Kim, Joongheon Kim
2020 Applied Sciences  
This study proposes a novel multi-module deep neural network framework which aims at improving intelligent long-term traffic forecasting.  ...  The performance of the framework is then evaluated for different cases, which include all plausible cases of driving, i.e., regular days, holidays, and days involving severe weather conditions.  ...  Author Contributions: S.R. and D.K. were the main researchers who initiated and organized the research work reported in the paper; S.R. and J.K. were responsible for building the final TW-DNN to improve  ... 
doi:10.3390/app10061938 fatcat:kbpbod6lrrhhlg2lu53avdfgvu

Deep Generative Quantile-Copula Models for Probabilistic Forecasting [article]

Ruofeng Wen, Kari Torkkola
2019 arXiv   pre-print
Then the multivariate case is solved by learning such quantile functions for each dimension's marginal distribution, followed by estimating a conditional Copula to associate these latent uniform random  ...  We introduce a new category of multivariate conditional generative models and demonstrate its performance and versatility in probabilistic time series forecasting and simulation.  ...  Specifically, we design a conditional generative Quantile-Copula framework, parameterized by a single deep neural network.  ... 
arXiv:1907.10697v1 fatcat:qmf3rnyjlzdrdp5dckmove7xyy

A Composite Quantile Fourier Neural Network for Multi-Step Probabilistic Forecasting of Nonstationary Univariate Time Series [article]

Kostas Hatalis, Shalinee Kishore
2020 arXiv   pre-print
We present a novel quantile Fourier neural network is for nonparametric probabilistic forecasting of univariate time series.  ...  This effectively is a form of extrapolation based nonlinear quantile regression applied for forecasting.  ...  Lastly, QFNN performs well when linear trend is present such as in wind power and short term stock prices.  ... 
arXiv:1712.09641v2 fatcat:c4yxrx3qgbe5nkrvoscsktocge
« Previous Showing results 1 — 15 out of 1,449 results