A New Look at Nonlinear Time Series Prediction with NARX Recurrent Neural Network. José Maria P. Menezes Jr. and Guilherme A.

Size: px
Start display at page:

Download "A New Look at Nonlinear Time Series Prediction with NARX Recurrent Neural Network. José Maria P. Menezes Jr. and Guilherme A."

Transcription

1 A New Look at Nonlinear Time Series Prediction with NARX Recurrent Neural Network José Maria P. Menezes Jr. and Guilherme A. Barreto Department of Teleinformatics Engineering Federal University of Ceará, Centro de Tecnologia Fortaleza-CE, Brazil October 23-27, 2006

2 Contents

3 Motivation Objectives Theoretical Foundations Time Series Prediction (TSP) Tasks Recurrent Neural Networks (RNN) NARX Architecture Simulations VBR Video Traffic Laser Time Series Conclusion

4 Motivation 1. Long Term Dependence occurs very often in real-world time series (e.g. traffic series). 2. Theory of Dynamical Systems provides the theoretical bases to analyzing nonlinear systems with chaotic behavior. 3. Recurrent Neural Networks are capable of representing arbitrary nonlinear dynamical mappings, such as those commonly found in nonlinear time series prediction. 4. NARX Model is a recurrent neural network capable of modeling efficiently time series with long-term dependences.

5 Objectives of this Work 1. To evaluate the performance of standard dynamic neural networks in difficult time series prediction tasks. 2. To propose a new field of application for NARX networks: prediction of univariate time series with long range dependencies.

6 Theoretical Foundations

7 Time Series Prediction (TSP) Tasks TSP One-step-ahead prediction: Neural network models are commonly used to estimating only the next value of a time series. Multi-step-ahead prediction: If the user is interested in a wider prediction horizon. The model s output should be fed back to the input regressor for a fixed but finite number of time steps. Dynamic modeling: If the prediction horizon tends to infinity, the neural network will act as an autonomous system, modeling the long-term dynamics of the system that generated the oberved time series.

8 Recurrent Neural Networks (RNN) RNN Feedforward MLP-like networks can be easily adapted to process time series through an input tapped delay line (e.g. FTDNN model). Recurrent neural networks (RNN) have local and/or global feedback loops in their structure (e.g. Elman, Jordan and NARX models) [1]. RNN are capable to represent arbitrary nonlinear dynamical mappings, such as those commonly found in nonlinear time series prediction tasks.

9 Recurrent Neural Networks (RNN) Takens Embedding Theorem Takens [3] has shown that the state of a deterministic dynamic system can be accurately reconstructed by a time window of finite length sliding over the observed time series as follows: where x 1 (n) [x(n) x(n τ) x(n (d E 1)τ)] T, x(n) is the value of the time series at time n, d E is the embedding dimension and τ is the embedding delay.

10 Recurrent Neural Networks (RNN) FTDNN Focused Time Delay Neural Network

11 Recurrent Neural Networks (RNN) Elman Network

12 NARX Architecture NARX Model in System Identification Nonlinear Autoregressive with exogenous input (NARX) [2]: y(n + 1) = f [y(n),..., y(n d y + 1); u(n), u(n 1),..., u(n d u + 1)]. = f [y(n); u(n)], where u(n) and y(n) denote, respectively, the input and the output of the model at discrete time n. The parameters d u 1 and d y 1, d u d y, are memory delays.

13 NARX Architecture NARX Neural Network Architecture

14 NARX Architecture NARX Network in Nonlinear Time Series Prediction Using Takens Theorem to build the input regressor: u(n) = [x(n) x(n τ) x(n (d E 1)τ)] T, where we set d u = d E. The output regressor y(n) can be written in two different modes, depending on the training modes of the NARX network: y p (n) = [ x(n),..., x(n d y + 1)], y sp (n) = [x(n),..., x(n d y + 1)], where the P-mode contains d y past values of the estimated time series, while the SP-mode contains d y past values of the actual time series.

15 NARX Architecture Paralell Mode (NARX-P)

16 NARX Architecture Series-Parallel Mode (NARX-SP)

17 Simulations

18 Evaluated Networks NARX-P, NARX-SP, FTDNN and Elman networks. All networks have two-hidden layers and one output neuron. All neurons use the hyperbolic tangent activation function. The standard backpropagation algorithm is used to train the networks.

19 Summary Table - Training Parameters number of neurons number of neurons 1st hidden layer 2st hidden layer learning rate epochs (N h,1 ) p (N h,2 ) TASK 1 2d E + 1 p Nh, TASK 2 2d E + 1 Nh,

20 Performance Evaluation Metric The networks are evaluated in multi-step-ahead prediction tasks. Quantitatively, we compute the Normalized Mean Squared Error (NMSE): NMSE(N) = 1 N N σx 2 e 2 (n) where n=1 N is the horizon prediction, σ x 2 is the sample variance of the actual time series and e(n) = y(n) ŷ(n) is the prediction error at time n.

21 Task 1: Long-term prediction of VBR video traffic Variable bit rate (VBR) video traffic (Jurassic Park) [4]. This video traffic trace was encoded with MPEG-I. VBR video traffic typically exhibits burstiness over multiple time scales [5],[6] sample points, rescaled to the range [ 1, 1] samples for training and 500 samples for testing.

22 VBR Video Traffic Empirical Sensitivity Analysis - 1 Embedding dimension 1 FTDNN Elman NARX P NARX SP 0.8 NMSE Order

23 VBR Video Traffic Empirical Sensitivity Analysis - 2 Number of training epochs FTDNN Elman NARX P NARX SP NMSE Epochs

24 VBR Video Traffic Multi-Step-Ahead Predictions - 1 FTDNN Predicted Original 0 Bits Frame number

25 VBR Video Traffic Multi-Step-Ahead Predictions - 2 Elman Predicted Original 0 Bits Frame number

26 VBR Video Traffic Multi-Step-Ahead Predictions - 3 NARX-SP 0.5 Predicted Original 0 Bits Frame number

27 VBR Video Traffic Task 2: Long-term prediction of chaotic laser intensities Chaotic laser time series: comprises measurements of the intensity pulsations of a single-mode Far-Infrared-Laser NH3 in a chaotic state [7]. Available worldwide since a TSP competition organized by the Santa Fe Institute [8] sample points which have been rescaled to the range [ 1, 1] samples for training and 500 samples for testing.

28 Laser Time Series Dynamic Modeling - 1 FTDNN Elman Network Predicted Original Predicted Original P 0 P Time Time

29 Laser Time Series Dynamic Modeling - 2 NARX-SP Network Predicted Original P Time

30 Laser Time Series Sensitivity Analysis Length of the prediction horizon FTDNN Elman NARX P NARX SP 2 Arv Prediction Horizon (N)

31 Contents Motivation Objectives Theoretical Foundations Simulations Conclusion Laser Time Series Laser Time Series j j Recurrence Plot: original series, NARX-SP, FTDNN, Elman i j j i i i

32 Conclusion

33 Conclusion The results has shown that NARX network can be succesfully applied to complex univariate time series modelling and prediction tasks. The proposed approach consistently outperforms standard neural network based predictors, such as the FTDNN and Elman architectures.

34 References [1] J. F. Kolen and S. C. Kremer, A Field Guide to Dynamical Recurrent Networks, Wiley-IEEE Press, [2] T. Lin, B. G. Horne, P. Tino, and C. L. Giles, Learning long-term dependencies in NARX recurrent neural networks IEEE Transactions on Neural Networks, vol. 7, no. 6, pp , [3] F. Takens, Detecting strange attractors in turbulence, in Dynamical Systems and Turbulence, D. A. Rand and L.-S. Young, Eds. 1981, vol. 898 of Lecture Notes in Mathematics, pp , Springer. [4] O. Rose, Statistical properties of MPEG video traffic and their impact on traffic modeling in ATM systems, in Proceedings of the 20th Annual IEEE Conference on Local. [5] J. Beran, R. Sherman, M. S. Taqqu, and W. Willinger, Long-range dependence in variable-bit-rate video traffic, IEEE Transactions on Communications, vol. 43, no. 234, pp , [6] D. Heyman and T. Lakshman, What are the implications of long-range dependence for VBR video traffic engineering, IEEE/ACM Transactions on Networking, vol. 4, no. 3, pp , [7] U. Huebner, N. B. Abraham, and C. O. Weiss. Dimensions and entropies of chaotic intensity pulsations in a single-mode far-infrared NH3 laser. Physical Review A, 40(11): , [8] A. Weigend and N. Gershefeld. Time Series Prediction: Forecasting the Future and Understanding the Past. Addison-Wesley, Reading, 1994.

35 A New Look at Nonlinear Time Series Prediction with NARX Recurrent Neural Network

Direct Method for Training Feed-forward Neural Networks using Batch Extended Kalman Filter for Multi- Step-Ahead Predictions

Direct Method for Training Feed-forward Neural Networks using Batch Extended Kalman Filter for Multi- Step-Ahead Predictions Direct Method for Training Feed-forward Neural Networks using Batch Extended Kalman Filter for Multi- Step-Ahead Predictions Artem Chernodub, Institute of Mathematical Machines and Systems NASU, Neurotechnologies

More information

MODELING NONLINEAR DYNAMICS WITH NEURAL. Eric A. Wan. Stanford University, Department of Electrical Engineering, Stanford, CA

MODELING NONLINEAR DYNAMICS WITH NEURAL. Eric A. Wan. Stanford University, Department of Electrical Engineering, Stanford, CA MODELING NONLINEAR DYNAMICS WITH NEURAL NETWORKS: EXAMPLES IN TIME SERIES PREDICTION Eric A Wan Stanford University, Department of Electrical Engineering, Stanford, CA 9435-455 Abstract A neural networ

More information

Reservoir Computing and Echo State Networks

Reservoir Computing and Echo State Networks An Introduction to: Reservoir Computing and Echo State Networks Claudio Gallicchio gallicch@di.unipi.it Outline Focus: Supervised learning in domain of sequences Recurrent Neural networks for supervised

More information

Lecture 5: Recurrent Neural Networks

Lecture 5: Recurrent Neural Networks 1/25 Lecture 5: Recurrent Neural Networks Nima Mohajerin University of Waterloo WAVE Lab nima.mohajerin@uwaterloo.ca July 4, 2017 2/25 Overview 1 Recap 2 RNN Architectures for Learning Long Term Dependencies

More information

y(n) Time Series Data

y(n) Time Series Data Recurrent SOM with Local Linear Models in Time Series Prediction Timo Koskela, Markus Varsta, Jukka Heikkonen, and Kimmo Kaski Helsinki University of Technology Laboratory of Computational Engineering

More information

Ensembles of Nearest Neighbor Forecasts

Ensembles of Nearest Neighbor Forecasts Ensembles of Nearest Neighbor Forecasts Dragomir Yankov 1, Dennis DeCoste 2, and Eamonn Keogh 1 1 University of California, Riverside CA 92507, USA, {dyankov,eamonn}@cs.ucr.edu, 2 Yahoo! Research, 3333

More information

T Machine Learning and Neural Networks

T Machine Learning and Neural Networks T-61.5130 Machine Learning and Neural Networks (5 cr) Lecture 11: Processing of Temporal Information Prof. Juha Karhunen https://mycourses.aalto.fi/ Aalto University School of Science, Espoo, Finland 1

More information

Chapter 15. Dynamically Driven Recurrent Networks

Chapter 15. Dynamically Driven Recurrent Networks Chapter 15. Dynamically Driven Recurrent Networks Neural Networks and Learning Machines (Haykin) Lecture Notes on Self-learning Neural Algorithms Byoung-Tak Zhang School of Computer Science and Engineering

More information

Information Dynamics Foundations and Applications

Information Dynamics Foundations and Applications Gustavo Deco Bernd Schürmann Information Dynamics Foundations and Applications With 89 Illustrations Springer PREFACE vii CHAPTER 1 Introduction 1 CHAPTER 2 Dynamical Systems: An Overview 7 2.1 Deterministic

More information

Long-Term Prediction, Chaos and Artificial Neural Networks. Where is the Meeting Point?

Long-Term Prediction, Chaos and Artificial Neural Networks. Where is the Meeting Point? Engineering Letters, 5:, EL_5 Long-Term Prediction, Chaos and Artificial Neural Networks. Where is the Meeting Point? Pilar Gómez-Gil Abstract This paper presents the advances of a research using a combination

More information

NARX neural networks for sequence processing tasks

NARX neural networks for sequence processing tasks Master in Artificial Intelligence (UPC-URV-UB) Master of Science Thesis NARX neural networks for sequence processing tasks eng. Eugen Hristev Advisor: prof. dr. René Alquézar Mancho June 2012 Table of

More information

A new method for short-term load forecasting based on chaotic time series and neural network

A new method for short-term load forecasting based on chaotic time series and neural network A new method for short-term load forecasting based on chaotic time series and neural network Sajjad Kouhi*, Navid Taghizadegan Electrical Engineering Department, Azarbaijan Shahid Madani University, Tabriz,

More information

TIME SERIES FORECASTING FOR OUTDOOR TEMPERATURE USING NONLINEAR AUTOREGRESSIVE NEURAL NETWORK MODELS

TIME SERIES FORECASTING FOR OUTDOOR TEMPERATURE USING NONLINEAR AUTOREGRESSIVE NEURAL NETWORK MODELS TIME SERIES FORECASTING FOR OUTDOOR TEMPERATURE USING NONLINEAR AUTOREGRESSIVE NEURAL NETWORK MODELS SANAM NAREJO,EROS PASERO Department of Electronics and Telecommunication, Politecnico Di Torino, Italy

More information

TECHNICAL RESEARCH REPORT

TECHNICAL RESEARCH REPORT TECHNICAL RESEARCH REPORT Reconstruction of Nonlinear Systems Using Delay Lines and Feedforward Networks by D.L. Elliott T.R. 95-17 ISR INSTITUTE FOR SYSTEMS RESEARCH Sponsored by the National Science

More information

Modeling and Predicting Chaotic Time Series

Modeling and Predicting Chaotic Time Series Chapter 14 Modeling and Predicting Chaotic Time Series To understand the behavior of a dynamical system in terms of some meaningful parameters we seek the appropriate mathematical model that captures the

More information

Christian Mohr

Christian Mohr Christian Mohr 20.12.2011 Recurrent Networks Networks in which units may have connections to units in the same or preceding layers Also connections to the unit itself possible Already covered: Hopfield

More information

Chapter 4 Neural Networks in System Identification

Chapter 4 Neural Networks in System Identification Chapter 4 Neural Networks in System Identification Gábor HORVÁTH Department of Measurement and Information Systems Budapest University of Technology and Economics Magyar tudósok körútja 2, 52 Budapest,

More information

Sample Exam COMP 9444 NEURAL NETWORKS Solutions

Sample Exam COMP 9444 NEURAL NETWORKS Solutions FAMILY NAME OTHER NAMES STUDENT ID SIGNATURE Sample Exam COMP 9444 NEURAL NETWORKS Solutions (1) TIME ALLOWED 3 HOURS (2) TOTAL NUMBER OF QUESTIONS 12 (3) STUDENTS SHOULD ANSWER ALL QUESTIONS (4) QUESTIONS

More information

Estimating the Number of Hidden Neurons of the MLP Using Singular Value Decomposition and Principal Components Analysis: A Novel Approach

Estimating the Number of Hidden Neurons of the MLP Using Singular Value Decomposition and Principal Components Analysis: A Novel Approach Estimating the Number of Hidden Neurons of the MLP Using Singular Value Decomposition and Principal Components Analysis: A Novel Approach José Daniel A Santos IFCE - Industry Department Av Contorno Norte,

More information

Negatively Correlated Echo State Networks

Negatively Correlated Echo State Networks Negatively Correlated Echo State Networks Ali Rodan and Peter Tiňo School of Computer Science, The University of Birmingham Birmingham B15 2TT, United Kingdom E-mail: {a.a.rodan, P.Tino}@cs.bham.ac.uk

More information

Autonomous learning algorithm for fully connected recurrent networks

Autonomous learning algorithm for fully connected recurrent networks Autonomous learning algorithm for fully connected recurrent networks Edouard Leclercq, Fabrice Druaux, Dimitri Lefebvre Groupe de Recherche en Electrotechnique et Automatique du Havre Université du Havre,

More information

International University Bremen Guided Research Proposal Improve on chaotic time series prediction using MLPs for output training

International University Bremen Guided Research Proposal Improve on chaotic time series prediction using MLPs for output training International University Bremen Guided Research Proposal Improve on chaotic time series prediction using MLPs for output training Aakash Jain a.jain@iu-bremen.de Spring Semester 2004 1 Executive Summary

More information

FORECASTING OF INFLATION IN BANGLADESH USING ANN MODEL

FORECASTING OF INFLATION IN BANGLADESH USING ANN MODEL FORECASTING OF INFLATION IN BANGLADESH USING ANN MODEL Rumana Hossain Department of Physical Science School of Engineering and Computer Science Independent University, Bangladesh Shaukat Ahmed Department

More information

Multivariable and Multiaxial Fatigue Life Assessment of Composite Materials using Neural Networks

Multivariable and Multiaxial Fatigue Life Assessment of Composite Materials using Neural Networks Multivariable and Multiaxial Fatigue Life Assessment of Composite Materials using Neural Networks Mas Irfan P. Hidayat Abstract In the present paper, multivariable and multiaxial fatigue life assessment

More information

Multi-Model Integration for Long-Term Time Series Prediction

Multi-Model Integration for Long-Term Time Series Prediction Multi-Model Integration for Long-Term Time Series Prediction Zifang Huang, Mei-Ling Shyu, James M. Tien Department of Electrical and Computer Engineering University of Miami, Coral Gables, FL, USA z.huang3@umiami.edu,

More information

Input Selection for Long-Term Prediction of Time Series

Input Selection for Long-Term Prediction of Time Series Input Selection for Long-Term Prediction of Time Series Jarkko Tikka, Jaakko Hollmén, and Amaury Lendasse Helsinki University of Technology, Laboratory of Computer and Information Science, P.O. Box 54,

More information

Experiments with a Hybrid-Complex Neural Networks for Long Term Prediction of Electrocardiograms

Experiments with a Hybrid-Complex Neural Networks for Long Term Prediction of Electrocardiograms IEEE. ransactions of the 6 International World Congress of Computational Intelligence, IJCNN 6 Experiments with a Hybrid-Complex Neural Networks for Long erm Prediction of Electrocardiograms Pilar Gómez-Gil,

More information

Deep Learning Architecture for Univariate Time Series Forecasting

Deep Learning Architecture for Univariate Time Series Forecasting CS229,Technical Report, 2014 Deep Learning Architecture for Univariate Time Series Forecasting Dmitry Vengertsev 1 Abstract This paper studies the problem of applying machine learning with deep architecture

More information

Elman Recurrent Neural Network in Thermal Modeling of Power Transformers

Elman Recurrent Neural Network in Thermal Modeling of Power Transformers Elman Recurrent Neural Network in Thermal Modeling of Power Transformers M. HELL, F. GOMIDE Department of Computer Engineering and Industrial Automation - DCA School of Electrical and Computer Engineering

More information

Recurrent Neural Networks

Recurrent Neural Networks Recurrent Neural Networks Datamining Seminar Kaspar Märtens Karl-Oskar Masing Today's Topics Modeling sequences: a brief overview Training RNNs with back propagation A toy example of training an RNN Why

More information

Forecasting Chaotic time series by a Neural Network

Forecasting Chaotic time series by a Neural Network Forecasting Chaotic time series by a Neural Network Dr. ATSALAKIS George Technical University of Crete, Greece atsalak@otenet.gr Dr. SKIADAS Christos Technical University of Crete, Greece atsalak@otenet.gr

More information

A STATE-SPACE NEURAL NETWORK FOR MODELING DYNAMICAL NONLINEAR SYSTEMS

A STATE-SPACE NEURAL NETWORK FOR MODELING DYNAMICAL NONLINEAR SYSTEMS A STATE-SPACE NEURAL NETWORK FOR MODELING DYNAMICAL NONLINEAR SYSTEMS Karima Amoura Patrice Wira and Said Djennoune Laboratoire CCSP Université Mouloud Mammeri Tizi Ouzou Algeria Laboratoire MIPS Université

More information

On the use of Long-Short Term Memory neural networks for time series prediction

On the use of Long-Short Term Memory neural networks for time series prediction On the use of Long-Short Term Memory neural networks for time series prediction Pilar Gómez-Gil National Institute of Astrophysics, Optics and Electronics ccc.inaoep.mx/~pgomez In collaboration with: J.

More information

MODULAR ECHO STATE NEURAL NETWORKS IN TIME SERIES PREDICTION

MODULAR ECHO STATE NEURAL NETWORKS IN TIME SERIES PREDICTION Computing and Informatics, Vol. 30, 2011, 321 334 MODULAR ECHO STATE NEURAL NETWORKS IN TIME SERIES PREDICTION Štefan Babinec, Jiří Pospíchal Department of Mathematics Faculty of Chemical and Food Technology

More information

Deep Recurrent Neural Networks

Deep Recurrent Neural Networks Deep Recurrent Neural Networks Artem Chernodub e-mail: a.chernodub@gmail.com web: http://zzphoto.me ZZ Photo IMMSP NASU 2 / 28 Neuroscience Biological-inspired models Machine Learning p x y = p y x p(x)/p(y)

More information

EE-559 Deep learning Recurrent Neural Networks

EE-559 Deep learning Recurrent Neural Networks EE-559 Deep learning 11.1. Recurrent Neural Networks François Fleuret https://fleuret.org/ee559/ Sun Feb 24 20:33:31 UTC 2019 Inference from sequences François Fleuret EE-559 Deep learning / 11.1. Recurrent

More information

Recurrent Neural Networks Deep Learning Lecture 5. Efstratios Gavves

Recurrent Neural Networks Deep Learning Lecture 5. Efstratios Gavves Recurrent Neural Networks Deep Learning Lecture 5 Efstratios Gavves Sequential Data So far, all tasks assumed stationary data Neither all data, nor all tasks are stationary though Sequential Data: Text

More information

Predicting the Future with the Appropriate Embedding Dimension and Time Lag JAMES SLUSS

Predicting the Future with the Appropriate Embedding Dimension and Time Lag JAMES SLUSS Predicting the Future with the Appropriate Embedding Dimension and Time Lag Georgios Lezos, Monte Tull, Joseph Havlicek, and Jim Sluss GEORGIOS LEZOS Graduate Student School of Electtical & Computer Engineering

More information

Echo State Networks with Filter Neurons and a Delay&Sum Readout

Echo State Networks with Filter Neurons and a Delay&Sum Readout Echo State Networks with Filter Neurons and a Delay&Sum Readout Georg Holzmann 2,1 (Corresponding Author) http://grh.mur.at grh@mur.at Helmut Hauser 1 helmut.hauser@igi.tugraz.at 1 Institute for Theoretical

More information

Learning Chaotic Attractors by Neural Networks

Learning Chaotic Attractors by Neural Networks LETTER Communicated by José Principe Learning Chaotic Attractors by Neural Networks Rembrandt Bakker DelftChemTech, Delft University of Technology, 2628 BL Delft, The Netherlands Jaap C. Schouten Chemical

More information

SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION. (An ANN Based Model using NARX Model for forecasting of GHI)

SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION. (An ANN Based Model using NARX Model for forecasting of GHI) IPS 2018 SOLAR ENERGY FORECASTING A PATHWAY FOR SUCCESSFUL RENEWABLE ENERGY INTEGRATION ABSTRACT (An ANN Based Model using NARX Model for forecasting of GHI) Manish Kumar Tikariha, DGM(Operations), NTPC

More information

Evaluating nonlinearity and validity of nonlinear modeling for complex time series

Evaluating nonlinearity and validity of nonlinear modeling for complex time series Evaluating nonlinearity and validity of nonlinear modeling for complex time series Tomoya Suzuki, 1 Tohru Ikeguchi, 2 and Masuo Suzuki 3 1 Department of Information Systems Design, Doshisha University,

More information

HALF HOURLY ELECTRICITY LOAD PREDICTION USING ECHO STATE NETWORK

HALF HOURLY ELECTRICITY LOAD PREDICTION USING ECHO STATE NETWORK HALF HOURLY ELECTRICITY LOAD PREDICTION USING ECHO STATE NETWORK Shivani Varshney, Toran Verma Department of Computer Science & Engineering, RCET, Bhilai, India. ABSTRACT Prediction of time series is a

More information

Financial Risk and Returns Prediction with Modular Networked Learning

Financial Risk and Returns Prediction with Modular Networked Learning arxiv:1806.05876v1 [cs.lg] 15 Jun 2018 Financial Risk and Returns Prediction with Modular Networked Learning Carlos Pedro Gonçalves June 18, 2018 University of Lisbon, Instituto Superior de Ciências Sociais

More information

Automatic modelling of neural networks for time series prediction in search of a uniform methodology across varying time frequencies

Automatic modelling of neural networks for time series prediction in search of a uniform methodology across varying time frequencies Automatic modelling of neural networks for time series prediction in search of a uniform methodology across varying time frequencies Nikolaos Kourentzes and Sven F. Crone Lancaster University Management

More information

Hybrid HMM/MLP models for time series prediction

Hybrid HMM/MLP models for time series prediction Bruges (Belgium), 2-23 April 999, D-Facto public., ISBN 2-649-9-X, pp. 455-462 Hybrid HMM/MLP models for time series prediction Joseph Rynkiewicz SAMOS, Université Paris I - Panthéon Sorbonne Paris, France

More information

Predicting Chaotic Time Series by Reinforcement Learning

Predicting Chaotic Time Series by Reinforcement Learning Predicting Chaotic Time Series by Reinforcement Learning T. Kuremoto 1, M. Obayashi 1, A. Yamamoto 1, and K. Kobayashi 1 1 Dep. of Computer Science and Systems Engineering, Engineering Faculty,Yamaguchi

More information

Short Term Memory Quantifications in Input-Driven Linear Dynamical Systems

Short Term Memory Quantifications in Input-Driven Linear Dynamical Systems Short Term Memory Quantifications in Input-Driven Linear Dynamical Systems Peter Tiňo and Ali Rodan School of Computer Science, The University of Birmingham Birmingham B15 2TT, United Kingdom E-mail: {P.Tino,

More information

NONLINEAR PLANT IDENTIFICATION BY WAVELETS

NONLINEAR PLANT IDENTIFICATION BY WAVELETS NONLINEAR PLANT IDENTIFICATION BY WAVELETS Edison Righeto UNESP Ilha Solteira, Department of Mathematics, Av. Brasil 56, 5385000, Ilha Solteira, SP, Brazil righeto@fqm.feis.unesp.br Luiz Henrique M. Grassi

More information

Experiments with neural network for modeling of nonlinear dynamical systems: Design problems

Experiments with neural network for modeling of nonlinear dynamical systems: Design problems Experiments with neural network for modeling of nonlinear dynamical systems: Design problems Ewa Skubalska-Rafaj lowicz Wroc law University of Technology, Wroc law, Wroc law, Poland Summary Introduction

More information

4. Multilayer Perceptrons

4. Multilayer Perceptrons 4. Multilayer Perceptrons This is a supervised error-correction learning algorithm. 1 4.1 Introduction A multilayer feedforward network consists of an input layer, one or more hidden layers, and an output

More information

System Identification for the Hodgkin-Huxley Model using Artificial Neural Networks

System Identification for the Hodgkin-Huxley Model using Artificial Neural Networks Proceedings of International Joint Conference on Neural Networks, Orlando, Florida, USA, August 12-17, 2007 System Identification for the Hodgkin-Huxley Model using Artificial Neural Networks Manish Saggar,

More information

A Black-Box Approach in Modeling Valve Stiction

A Black-Box Approach in Modeling Valve Stiction Vol:4, No:8, A Black-Box Approach in Modeling Valve Stiction H. Zabiri, N. Mazuki International Science Index, Mechanical and Mechatronics Engineering Vol:4, No:8, waset.org/publication/46 Abstract Several

More information

Analysis of Fast Input Selection: Application in Time Series Prediction

Analysis of Fast Input Selection: Application in Time Series Prediction Analysis of Fast Input Selection: Application in Time Series Prediction Jarkko Tikka, Amaury Lendasse, and Jaakko Hollmén Helsinki University of Technology, Laboratory of Computer and Information Science,

More information

Neural Network Identification of Non Linear Systems Using State Space Techniques.

Neural Network Identification of Non Linear Systems Using State Space Techniques. Neural Network Identification of Non Linear Systems Using State Space Techniques. Joan Codina, J. Carlos Aguado, Josep M. Fuertes. Automatic Control and Computer Engineering Department Universitat Politècnica

More information

Recurrent Neural Networks (Part - 2) Sumit Chopra Facebook

Recurrent Neural Networks (Part - 2) Sumit Chopra Facebook Recurrent Neural Networks (Part - 2) Sumit Chopra Facebook Recap Standard RNNs Training: Backpropagation Through Time (BPTT) Application to sequence modeling Language modeling Applications: Automatic speech

More information

Analysis of Multilayer Neural Network Modeling and Long Short-Term Memory

Analysis of Multilayer Neural Network Modeling and Long Short-Term Memory Analysis of Multilayer Neural Network Modeling and Long Short-Term Memory Danilo López, Nelson Vera, Luis Pedraza International Science Index, Mathematical and Computational Sciences waset.org/publication/10006216

More information

Sensitivity of ABR Congestion Control Algorithms to Hurst Parameter Estimates

Sensitivity of ABR Congestion Control Algorithms to Hurst Parameter Estimates Sensitivity of ABR Congestion Control Algorithms to Hurst Parameter Estimates Sven A. M. Östring 1, Harsha Sirisena 1, and Irene Hudson 2 1 Department of Electrical & Electronic Engineering 2 Department

More information

PHONEME CLASSIFICATION OVER THE RECONSTRUCTED PHASE SPACE USING PRINCIPAL COMPONENT ANALYSIS

PHONEME CLASSIFICATION OVER THE RECONSTRUCTED PHASE SPACE USING PRINCIPAL COMPONENT ANALYSIS PHONEME CLASSIFICATION OVER THE RECONSTRUCTED PHASE SPACE USING PRINCIPAL COMPONENT ANALYSIS Jinjin Ye jinjin.ye@mu.edu Michael T. Johnson mike.johnson@mu.edu Richard J. Povinelli richard.povinelli@mu.edu

More information

Modelling Time Series with Neural Networks. Volker Tresp Summer 2017

Modelling Time Series with Neural Networks. Volker Tresp Summer 2017 Modelling Time Series with Neural Networks Volker Tresp Summer 2017 1 Modelling of Time Series The next figure shows a time series (DAX) Other interesting time-series: energy prize, energy consumption,

More information

Transformer Top-Oil Temperature Modeling and Simulation

Transformer Top-Oil Temperature Modeling and Simulation Transformer Top-Oil Temperature Modeling and Simulation T. C. B. N. Assunção, J. L. Silvino, and P. Resende Abstract The winding hot-spot temperature is one of the most critical parameters that affect

More information

Introduction to Neural Networks: Structure and Training

Introduction to Neural Networks: Structure and Training Introduction to Neural Networks: Structure and Training Professor Q.J. Zhang Department of Electronics Carleton University, Ottawa, Canada www.doe.carleton.ca/~qjz, qjz@doe.carleton.ca A Quick Illustration

More information

Load Forecasting Using Artificial Neural Networks and Support Vector Regression

Load Forecasting Using Artificial Neural Networks and Support Vector Regression Proceedings of the 7th WSEAS International Conference on Power Systems, Beijing, China, September -7, 2007 3 Load Forecasting Using Artificial Neural Networks and Support Vector Regression SILVIO MICHEL

More information

Reconstruction Deconstruction:

Reconstruction Deconstruction: Reconstruction Deconstruction: A Brief History of Building Models of Nonlinear Dynamical Systems Jim Crutchfield Center for Computational Science & Engineering Physics Department University of California,

More information

Dynamical Systems and Deep Learning: Overview. Abbas Edalat

Dynamical Systems and Deep Learning: Overview. Abbas Edalat Dynamical Systems and Deep Learning: Overview Abbas Edalat Dynamical Systems The notion of a dynamical system includes the following: A phase or state space, which may be continuous, e.g. the real line,

More information

Memory Capacity of Input-Driven Echo State NetworksattheEdgeofChaos

Memory Capacity of Input-Driven Echo State NetworksattheEdgeofChaos Memory Capacity of Input-Driven Echo State NetworksattheEdgeofChaos Peter Barančok and Igor Farkaš Faculty of Mathematics, Physics and Informatics Comenius University in Bratislava, Slovakia farkas@fmph.uniba.sk

More information

Neural Networks with Applications to Vision and Language. Feedforward Networks. Marco Kuhlmann

Neural Networks with Applications to Vision and Language. Feedforward Networks. Marco Kuhlmann Neural Networks with Applications to Vision and Language Feedforward Networks Marco Kuhlmann Feedforward networks Linear separability x 2 x 2 0 1 0 1 0 0 x 1 1 0 x 1 linearly separable not linearly separable

More information

MODELLING OF METALLURGICAL PROCESSES USING CHAOS THEORY AND HYBRID COMPUTATIONAL INTELLIGENCE

MODELLING OF METALLURGICAL PROCESSES USING CHAOS THEORY AND HYBRID COMPUTATIONAL INTELLIGENCE MODELLING OF METALLURGICAL PROCESSES USING CHAOS THEORY AND HYBRID COMPUTATIONAL INTELLIGENCE J. Krishanaiah, C. S. Kumar, M. A. Faruqi, A. K. Roy Department of Mechanical Engineering, Indian Institute

More information

Control-oriented model learning with a recurrent neural network

Control-oriented model learning with a recurrent neural network Control-oriented model learning with a recurrent neural network M. A. Bucci O. Semeraro A. Allauzen L. Cordier G. Wisniewski L. Mathelin 20 November 2018, APS Atlanta Kuramoto-Sivashinsky (KS) u t = 4

More information

USING WAVELET NEURAL NETWORK FOR THE IDENTIFICATION OF A BUILDING STRUCTURE FROM EXPERIMENTAL DATA

USING WAVELET NEURAL NETWORK FOR THE IDENTIFICATION OF A BUILDING STRUCTURE FROM EXPERIMENTAL DATA 13 th World Conference on Earthquake Engineering Vancouver, B.C., Canada August 1-6, 24 Paper No. 241 USING WAVELET NEURAL NETWORK FOR THE IDENTIFICATION OF A BUILDING STRUCTURE FROM EXPERIMENTAL DATA

More information

Artificial Neural Network

Artificial Neural Network Artificial Neural Network Eung Je Woo Department of Biomedical Engineering Impedance Imaging Research Center (IIRC) Kyung Hee University Korea ejwoo@khu.ac.kr Neuron and Neuron Model McCulloch and Pitts

More information

Analysis of the Learning Process of a Recurrent Neural Network on the Last k-bit Parity Function

Analysis of the Learning Process of a Recurrent Neural Network on the Last k-bit Parity Function Analysis of the Learning Process of a Recurrent Neural Network on the Last k-bit Parity Function Austin Wang Adviser: Xiuyuan Cheng May 4, 2017 1 Abstract This study analyzes how simple recurrent neural

More information

Wavelet Neural Networks for Nonlinear Time Series Analysis

Wavelet Neural Networks for Nonlinear Time Series Analysis Applied Mathematical Sciences, Vol. 4, 2010, no. 50, 2485-2495 Wavelet Neural Networks for Nonlinear Time Series Analysis K. K. Minu, M. C. Lineesh and C. Jessy John Department of Mathematics National

More information

(Feed-Forward) Neural Networks Dr. Hajira Jabeen, Prof. Jens Lehmann

(Feed-Forward) Neural Networks Dr. Hajira Jabeen, Prof. Jens Lehmann (Feed-Forward) Neural Networks 2016-12-06 Dr. Hajira Jabeen, Prof. Jens Lehmann Outline In the previous lectures we have learned about tensors and factorization methods. RESCAL is a bilinear model for

More information

EEE 241: Linear Systems

EEE 241: Linear Systems EEE 4: Linear Systems Summary # 3: Introduction to artificial neural networks DISTRIBUTED REPRESENTATION An ANN consists of simple processing units communicating with each other. The basic elements of

More information

This paper presents the

This paper presents the ISESCO JOURNAL of Science and Technology Volume 8 - Number 14 - November 2012 (2-8) A Novel Ensemble Neural Network based Short-term Wind Power Generation Forecasting in a Microgrid Aymen Chaouachi and

More information

Recurrent Neural Network Based Gating for Natural Gas Load Prediction System

Recurrent Neural Network Based Gating for Natural Gas Load Prediction System Recurrent Neural Network Based Gating for Natural Gas Load Prediction System Petr Musilek, Member, IEEE, Emil Pelikán, Tomáš Brabec and Milan Šimůnek Abstract Prediction of natural gas consumption is an

More information

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting 卷积 LSTM 网络 : 利用机器学习预测短期降雨 施行健 香港科技大学 VALSE 2016/03/23

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting 卷积 LSTM 网络 : 利用机器学习预测短期降雨 施行健 香港科技大学 VALSE 2016/03/23 Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting 卷积 LSTM 网络 : 利用机器学习预测短期降雨 施行健 香港科技大学 VALSE 2016/03/23 Content Quick Review of Recurrent Neural Network Introduction

More information

Fractional Integrated Recurrent Neural Network. (FIRNN) for Forecasting of Time Series Data in. Electricity Load in Java-Bali

Fractional Integrated Recurrent Neural Network. (FIRNN) for Forecasting of Time Series Data in. Electricity Load in Java-Bali Contemporary Engineering Sciences, Vol. 8, 2015, no. 32, 1535-1550 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ces.2015.510283 Fractional Integrated Recurrent Neural Network (FIRNN) for Forecasting

More information

Modeling Economic Time Series Using a Focused Time Lagged FeedForward Neural Network

Modeling Economic Time Series Using a Focused Time Lagged FeedForward Neural Network Proceedings of Student Research Day, CSIS, Pace University, May 9th, 23 Modeling Economic Time Series Using a Focused Time Lagged FeedForward Neural Network N. Moseley ABSTRACT, - Artificial neural networks

More information

Lecture 15: Exploding and Vanishing Gradients

Lecture 15: Exploding and Vanishing Gradients Lecture 15: Exploding and Vanishing Gradients Roger Grosse 1 Introduction Last lecture, we introduced RNNs and saw how to derive the gradients using backprop through time. In principle, this lets us train

More information

Introduction to Neural Networks

Introduction to Neural Networks Introduction to Neural Networks What are (Artificial) Neural Networks? Models of the brain and nervous system Highly parallel Process information much more like the brain than a serial computer Learning

More information

Nonlinear Characterization of Activity Dynamics in Online Collaboration Websites

Nonlinear Characterization of Activity Dynamics in Online Collaboration Websites Nonlinear Characterization of Activity Dynamics in Online Collaboration Websites Tiago Santos 1 Simon Walk 2 Denis Helic 3 1 Know-Center, Graz, Austria 2 Stanford University 3 Graz University of Technology

More information

A recursive algorithm based on the extended Kalman filter for the training of feedforward neural models. Isabelle Rivals and Léon Personnaz

A recursive algorithm based on the extended Kalman filter for the training of feedforward neural models. Isabelle Rivals and Léon Personnaz In Neurocomputing 2(-3): 279-294 (998). A recursive algorithm based on the extended Kalman filter for the training of feedforward neural models Isabelle Rivals and Léon Personnaz Laboratoire d'électronique,

More information

The Research of Railway Coal Dispatched Volume Prediction Based on Chaos Theory

The Research of Railway Coal Dispatched Volume Prediction Based on Chaos Theory The Research of Railway Coal Dispatched Volume Prediction Based on Chaos Theory Hua-Wen Wu Fu-Zhang Wang Institute of Computing Technology, China Academy of Railway Sciences Beijing 00044, China, P.R.

More information

Multilayer Perceptrons (MLPs)

Multilayer Perceptrons (MLPs) CSE 5526: Introduction to Neural Networks Multilayer Perceptrons (MLPs) 1 Motivation Multilayer networks are more powerful than singlelayer nets Example: XOR problem x 2 1 AND x o x 1 x 2 +1-1 o x x 1-1

More information

A Logarithmic Neural Network Architecture for Unbounded Non-Linear Function Approximation

A Logarithmic Neural Network Architecture for Unbounded Non-Linear Function Approximation 1 Introduction A Logarithmic Neural Network Architecture for Unbounded Non-Linear Function Approximation J Wesley Hines Nuclear Engineering Department The University of Tennessee Knoxville, Tennessee,

More information

Learning and Memory in Neural Networks

Learning and Memory in Neural Networks Learning and Memory in Neural Networks Guy Billings, Neuroinformatics Doctoral Training Centre, The School of Informatics, The University of Edinburgh, UK. Neural networks consist of computational units

More information

Pattern Matching and Neural Networks based Hybrid Forecasting System

Pattern Matching and Neural Networks based Hybrid Forecasting System Pattern Matching and Neural Networks based Hybrid Forecasting System Sameer Singh and Jonathan Fieldsend PA Research, Department of Computer Science, University of Exeter, Exeter, UK Abstract In this paper

More information

Power and Limits of Recurrent Neural Networks for Symbolic Sequences Processing

Power and Limits of Recurrent Neural Networks for Symbolic Sequences Processing Power and Limits of Recurrent Neural Networks for Symbolic Sequences Processing Matej Makula Institute of Applied Informatics Faculty of Informatics and Information Technologies Slovak University of Technology

More information

Application of an Artificial Neural Network Based Tool for Prediction of Pavement Performance

Application of an Artificial Neural Network Based Tool for Prediction of Pavement Performance 0 0 0 0 Application of an Artificial Neural Network Based Tool for Prediction of Pavement Performance Adelino Ferreira, Rodrigo Cavalcante Pavement Mechanics Laboratory, Research Center for Territory,

More information

Recurrent neural networks

Recurrent neural networks 12-1: Recurrent neural networks Prof. J.C. Kao, UCLA Recurrent neural networks Motivation Network unrollwing Backpropagation through time Vanishing and exploding gradients LSTMs GRUs 12-2: Recurrent neural

More information

Robust Learning of Chaotic Attractors

Robust Learning of Chaotic Attractors published in: Advances in Neural Information Processing Systems 12, S.A. Solla, T.K. Leen, K.-R. Müller (eds.), MIT Press, 2000, pp. 879--885. Robust Learning of Chaotic Attractors Rembrandt Bakker* Jaap

More information

Neural Network Ensemble-based Solar Power Generation Short-Term Forecasting

Neural Network Ensemble-based Solar Power Generation Short-Term Forecasting World Academy of Science, Engineering and Technology 5 9 eural etwork Ensemble-based Solar Power Generation Short-Term Forecasting A. Chaouachi, R. M. Kamel, R. Ichikawa, H. Hayashi, and K. agasaka Abstract

More information

Application of NARX based FFNN, SVR and ANN Fitting models for long term industrial load forecasting and their comparison

Application of NARX based FFNN, SVR and ANN Fitting models for long term industrial load forecasting and their comparison Application of NARX based FFNN, SVR and ANN Fitting models for long term industrial load forecasting and their comparison *Shahid M. Awan 1, 3, Member, IEEE, Zubair. A. Khan 1, 2, M. Aslam 3, Waqar Mahmood

More information

Automatic Structure and Parameter Training Methods for Modeling of Mechanical System by Recurrent Neural Networks

Automatic Structure and Parameter Training Methods for Modeling of Mechanical System by Recurrent Neural Networks Automatic Structure and Parameter Training Methods for Modeling of Mechanical System by Recurrent Neural Networks C. James Li and Tung-Yung Huang Department of Mechanical Engineering, Aeronautical Engineering

More information

Error Entropy Criterion in Echo State Network Training

Error Entropy Criterion in Echo State Network Training Error Entropy Criterion in Echo State Network Training Levy Boccato 1, Daniel G. Silva 1, Denis Fantinato 1, Kenji Nose Filho 1, Rafael Ferrari 1, Romis Attux 1, Aline Neves 2, Jugurta Montalvão 3 and

More information

NARX Time Series Model for Remaining Useful Life Estimation of Gas Turbine Engines

NARX Time Series Model for Remaining Useful Life Estimation of Gas Turbine Engines NARX Series Model for Remaining Useful Life Estimation of Gas Turbine Engines Oguz Bektas, Jeffrey A. Jones 2,2 Warwick Manufacturing Group, University of Warwick, Coventry, UK O.Bektas@warwick.ac.uk J.A.Jones@warwick.ac.uk

More information

Do we need Experts for Time Series Forecasting?

Do we need Experts for Time Series Forecasting? Do we need Experts for Time Series Forecasting? Christiane Lemke and Bogdan Gabrys Bournemouth University - School of Design, Engineering and Computing Poole House, Talbot Campus, Poole, BH12 5BB - United

More information

Long-Short Term Memory and Other Gated RNNs

Long-Short Term Memory and Other Gated RNNs Long-Short Term Memory and Other Gated RNNs Sargur Srihari srihari@buffalo.edu This is part of lecture slides on Deep Learning: http://www.cedar.buffalo.edu/~srihari/cse676 1 Topics in Sequence Modeling

More information