SlideShare uma empresa Scribd logo
1 de 29
Baixar para ler offline
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Time-series forecasting of indoor temperature
using pre-trained Deep Neural Networks
P. Romeu, F. Zamora-Mart´ınez, P. Botella-Rocamora, J. Pardo
Embedded Systems and Artificial Intelligence group
Departamento de ciencias f´ısicas, matem´aticas y de la computaci´on
Escuela Superior de Ense˜nanzas T´ecnicas (ESET)
Universidad CEU Cardenal Herrera, 46115 Alfara del Patriarca, Valencia (Spain)
ICANN – September 11, 2013
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Introduction and motivation
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Introduction and motivation
Introduction and motivation
Time series forecasting: prediction future values given past data.
¯s = s0,...,si−1,si,si+1,...
Non-linear relationships could be found between the elements.
ANNs were widely used for this task, normally shallow models.
Deep architectures has been successful in computer vision,
speech signal processing, classification, . . .
Time series forecasting with deep architectures is starting to
receive interest (as far as we know, using Restricted Boltzmann
Machines).
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Introduction and motivation
Deep architectures on time series
Expectations
Time series are characterized by more or less complex
dependencies. For indoor temperature forecasting:
Known dependencies: time of the day, day of the year.
Hidden dependencies: number of people in a room.
Short-term dependencies and long-term dependencies.
Normally, expert knowledge is introduced to take into account
known dependencies; data preprocessing: detrend, deseasoned.
A deep model could learn some of these dependencies using
several layers.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Introduction and motivation
Forecasting of indoor temperature with deep ANNs
What have we done in this work?
Evaluation of pre-training and denoising techniques in a time
series forecasting task.
Results: slightly better generalization, less over-fitting.
Problems: lack of data, not complex enough task.
15
16
17
18
19
20
21
22
23
24
25
26
0 2000 4000 6000 8000 10000
ºC
Time (minutes)
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Stacked Denoising Auto-Encoders
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Stacked Denoising Auto-Encoders
Stacked Denoising Auto-Encoders
A Denoising Auto-Encoder is a neural network which receives a
noisy input and produces its cleaned version.
Gaussian additive noise (σ): ˙x = x+N (0,σ2
I)
Masking noise (p): ˜x = MN(˙x) with p probability.
Encoding: h(˜x) = so ftsign(b+W ˜x)
Decoding (denoising): ˆx = g(h(˜x)) = so ftsign(c+WT
h(˜x))
˙x
h(˜x)
˜x ˆx
W WT
x
GN(x)
MN(˙x)
x is an input vector, h(·) is the hidden layer vector, b and c are bias
vectors, W is a weights matrix, softsign(·) =
x
1+|x|
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Stacked Denoising Auto-Encoders
Stacked Denoising Auto-Encoders
Greedy training building layer-by-layer auto-encoders.
Stack all the trained weights to produce the final result.
Stack a forecasting layer (linear activation).
Train the whole neural network.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Time series forecasting
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Time series forecasting
Time series forecasting
Univariate vs multivariate.
Single-step-ahead vs multi-step-ahead.
Iterative forecasting vs direct forecasting.
Multiple Input One Output vs Multiple Input Multiple Output.
ˆst+H
t+1 = F(st
t−I+1)
MIMO modelling is natural in ANNs, because they take profit of the
input/output mapping.
F is a forecasting model, H the number of predicted samples, I the number of past
samples taken as input.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Dataset
Dataset
Captured during 2011, March
and June.
1 minute sampling period.
Reduced and smoothed by
computing mean every 15
samples.
Differences between adjacent
samples were computed to
remove the trend.
Partition # of samples # of days
Training 2016 21
Validation 672 7
Test 672 7
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Evaluation measures
Evaluation measures
Mean Absolute Error (MAE)
Root Mean Square Error (RMSE)
MAE (t) =
1
|D|
|D|
∑
t=I
1
H
H
∑
h=1
|ˆst+h −st+h|
RMSE (t) =
1
|D|
|D|
∑
t=I
1
H
H
∑
h=1
(ˆst+h −st+h)2
|D| is the size of the dataset, H the future horizon, ˆst+h the forecasted value, st+h the
ground truth.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Experiments
Experiments
Different training modes comparison
TM-0 consists in a standard training of an ANN.
TM-1 pre-train the ANN using SDAE and fine-tuning of the whole
network
TM-2 pre-train the ANN using SDAE and fine-tuning of only last
layer (forecasting layer).
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Experiments
Experiments
Training description
Back-propagation with mini-batch size 32.
Mean Square Error (MSE) loss function.
Future horizon of 12 samples (three hours).
Minimum of 50 epochs, maximum of 4000.
Random search hyper-parameter optimization:
learning rate, momentum, weight decay,
number of hidden layers, hidden layer sizes,
number of inputs,
mask noise percentage.
3600 experiments for tuning.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Results
Results
Best topologies
- TM-0: 60 — 756 — 60 — 12
- TM-1: 48 — 648 — 920 — 16 — 12
- TM-2: 96 — 712 — 12
TM-0 has convergence problems with deep networks:
33% of two layered network experiments do not converge.
58% of three layered network experiments do not converge.
Note that the topologies are not the same in the three cases, we took the best
topology for each training mode.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Results
Results
20 random initializations of best hyper-parameters
0.115
0.120
0.125
0.130
0.135
0.140
TM-0 TM-1 TM-2
MAE*
Validation
Test
0.135
0.140
0.145
0.150
0.155
0.160
0.165
0.170
TM-0 TM-1 TM-2
RMSE*
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Results
Results
MSE of training partition during training
0.010
0.014
0.019
0.025
0.034
0.046
0.063
0.086
0.117
0.159
0 200 400 600 800 1000 1200 1400
TrainingMSE(log-scaled)
Epochs
TM-0
TM-1
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Experimentation
Results
Results
MAE of test partition during training
0.117
0.159
0.216
0.293
0.398
0 200 400 600 800 1000 1200 1400
TestMAE*(log-scaled)
Epochs
best val TM-0
best val TM-1
best val TM-2
TM-0
TM-1
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Conclusions and future work
Index
1 Introduction and motivation
2 Stacked Denoising Auto-Encoders
3 Time series forecasting
4 Experimentation
5 Conclusions and future work
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Conclusions and future work
Conclusions and future work
Pre-training, denoising techniques, and random hyper-parameter
optimization were used to carry out deep ANNs training in a
forecasting task.
Slightly better generalization performance at test set and a
reduction in over-fitting was observed (TM-1).
Fine-tuning phase of the whole deep model was needed to
ensure good results (TM-1 vs TM-2).
The short benefit of SDAE could be due to the low dimensionality
of the task.
In the future, this work will be extended by using larger
forecasting input window combined with multivariate forecasting.
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Conclusions and future work
Questions?
Thanks for your attention!
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: Hyper-parameter optimization
Grid search part
Train Mode: TM-0, TM-1, TM-2
Number of hidden layers: 1, 2, 3
Mask Noise: 0.02, 0.04, 0.10, 0.20
Random search part
100 random trials for every grid sweep
Input size: 12, 24, 36, 48, 60, 72, 84, 96
Learning rate: [10−3
,10−2
]
Momentum: ∼ N (10−3
,5×10−3
), ignoring negative values
Weight decay: [0,10−5
]
Hidden layer sizes: [4,1024]
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: hyper-parameters analysis
Input size
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.20
0.21
12 36 60 84
TM-0
12 36 60 84
TM-1
1 layer
2 layers
3 layers
12 36 60 84
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: hyper-parameters analysis
Encoding layer size
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.20
0.21
0 300 600 900
TM-0
0 300 600 900
TM-1
0 300 600 900
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: hyper-parameters analysis
Masking noise
0.12
0.13
0.14
0.15
0.16
0.17
0.02 0.10 0.18
TM-0
0.02 0.10 0.18
TM-1
0.02 0.10 0.18
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: hyper-parameters analysis
Learning rate of forecasting phase
0.12
0.13
0.14
0.15
0.16
0.17
0 0.003 0.006 0.009
TM-0
0 0.003 0.006 0.009
TM-1
0 0.003 0.006 0.009
TM-2
Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks
Appendix
Appendix: results table
MAE
Validation (µ±σ) Test (µ±σ)
ETS 0.3004 0.3254
TM-0 0.1289±0.0011 0.12482±0.0010
TM-1 0.1287±0.0033 0.1223±0.0033
TM-2 0.1374±0.0007 0.1279±0.0011
RMSE
Validation (µ±σ) Test (µ±σ)
ETS 0.3648 0.3930
TM-0 0.1563±0.0011 0.1511±0.0012
TM-1 0.1565±0.0040 0.1473±0.0039
TM-2 0.1663±0.0009 0.1538±0.0013

Mais conteúdo relacionado

Mais procurados

TensorFlow Tutorial Part1
TensorFlow Tutorial Part1TensorFlow Tutorial Part1
TensorFlow Tutorial Part1Sungjoon Choi
 
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...Taegyun Jeon
 
RNN & LSTM: Neural Network for Sequential Data
RNN & LSTM: Neural Network for Sequential DataRNN & LSTM: Neural Network for Sequential Data
RNN & LSTM: Neural Network for Sequential DataYao-Chieh Hu
 
Convolutional Neural Network
Convolutional Neural NetworkConvolutional Neural Network
Convolutional Neural NetworkJunho Cho
 
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)Universitat Politècnica de Catalunya
 
Recurrent neural networks for sequence learning and learning human identity f...
Recurrent neural networks for sequence learning and learning human identity f...Recurrent neural networks for sequence learning and learning human identity f...
Recurrent neural networks for sequence learning and learning human identity f...SungminYou
 
End-to-End Object Detection with Transformers
End-to-End Object Detection with TransformersEnd-to-End Object Detection with Transformers
End-to-End Object Detection with TransformersSeunghyun Hwang
 
Recurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRURecurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRUananth
 
Introduction to deep learning in python and Matlab
Introduction to deep learning in python and MatlabIntroduction to deep learning in python and Matlab
Introduction to deep learning in python and MatlabImry Kissos
 
CNNs: from the Basics to Recent Advances
CNNs: from the Basics to Recent AdvancesCNNs: from the Basics to Recent Advances
CNNs: from the Basics to Recent AdvancesDmytro Mishkin
 
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...Universitat Politècnica de Catalunya
 
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...Myungyon Kim
 
TypeScript and Deep Learning
TypeScript and Deep LearningTypeScript and Deep Learning
TypeScript and Deep LearningOswald Campesato
 
Introduction to Tree-LSTMs
Introduction to Tree-LSTMsIntroduction to Tree-LSTMs
Introduction to Tree-LSTMsDaniel Perez
 
A Brief Introduction on Recurrent Neural Network and Its Application
A Brief Introduction on Recurrent Neural Network and Its ApplicationA Brief Introduction on Recurrent Neural Network and Its Application
A Brief Introduction on Recurrent Neural Network and Its ApplicationXiaohu ZHU
 
Supervised sequence labelling with recurrent neural networks ch1 6
Supervised sequence labelling with recurrent neural networks ch1 6Supervised sequence labelling with recurrent neural networks ch1 6
Supervised sequence labelling with recurrent neural networks ch1 6SungminYou
 
Fundamental of deep learning
Fundamental of deep learningFundamental of deep learning
Fundamental of deep learningStanley Wang
 
Convolutional neural networks for image classification — evidence from Kaggle...
Convolutional neural networks for image classification — evidence from Kaggle...Convolutional neural networks for image classification — evidence from Kaggle...
Convolutional neural networks for image classification — evidence from Kaggle...Dmytro Mishkin
 
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Task
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” TaskThe Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Task
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Taskmultimediaeval
 

Mais procurados (19)

TensorFlow Tutorial Part1
TensorFlow Tutorial Part1TensorFlow Tutorial Part1
TensorFlow Tutorial Part1
 
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...
[PR12] PR-050: Convolutional LSTM Network: A Machine Learning Approach for Pr...
 
RNN & LSTM: Neural Network for Sequential Data
RNN & LSTM: Neural Network for Sequential DataRNN & LSTM: Neural Network for Sequential Data
RNN & LSTM: Neural Network for Sequential Data
 
Convolutional Neural Network
Convolutional Neural NetworkConvolutional Neural Network
Convolutional Neural Network
 
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)
Deep Learning for Computer Vision: Recurrent Neural Networks (UPC 2016)
 
Recurrent neural networks for sequence learning and learning human identity f...
Recurrent neural networks for sequence learning and learning human identity f...Recurrent neural networks for sequence learning and learning human identity f...
Recurrent neural networks for sequence learning and learning human identity f...
 
End-to-End Object Detection with Transformers
End-to-End Object Detection with TransformersEnd-to-End Object Detection with Transformers
End-to-End Object Detection with Transformers
 
Recurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRURecurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRU
 
Introduction to deep learning in python and Matlab
Introduction to deep learning in python and MatlabIntroduction to deep learning in python and Matlab
Introduction to deep learning in python and Matlab
 
CNNs: from the Basics to Recent Advances
CNNs: from the Basics to Recent AdvancesCNNs: from the Basics to Recent Advances
CNNs: from the Basics to Recent Advances
 
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...
Recurrent Neural Networks (DLAI D7L1 2017 UPC Deep Learning for Artificial In...
 
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...
Deep Learning and Tensorflow Implementation(딥러닝, 텐서플로우, 파이썬, CNN)_Myungyon Ki...
 
TypeScript and Deep Learning
TypeScript and Deep LearningTypeScript and Deep Learning
TypeScript and Deep Learning
 
Introduction to Tree-LSTMs
Introduction to Tree-LSTMsIntroduction to Tree-LSTMs
Introduction to Tree-LSTMs
 
A Brief Introduction on Recurrent Neural Network and Its Application
A Brief Introduction on Recurrent Neural Network and Its ApplicationA Brief Introduction on Recurrent Neural Network and Its Application
A Brief Introduction on Recurrent Neural Network and Its Application
 
Supervised sequence labelling with recurrent neural networks ch1 6
Supervised sequence labelling with recurrent neural networks ch1 6Supervised sequence labelling with recurrent neural networks ch1 6
Supervised sequence labelling with recurrent neural networks ch1 6
 
Fundamental of deep learning
Fundamental of deep learningFundamental of deep learning
Fundamental of deep learning
 
Convolutional neural networks for image classification — evidence from Kaggle...
Convolutional neural networks for image classification — evidence from Kaggle...Convolutional neural networks for image classification — evidence from Kaggle...
Convolutional neural networks for image classification — evidence from Kaggle...
 
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Task
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” TaskThe Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Task
The Munich LSTM-RNN Approach to the MediaEval 2014 “Emotion in Music” Task
 

Semelhante a Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks

Ann model and its application
Ann model and its applicationAnn model and its application
Ann model and its applicationmilan107
 
A Study on Privacy Level in Publishing Data of Smart Tap Network
A Study on Privacy Level in Publishing Data of Smart Tap NetworkA Study on Privacy Level in Publishing Data of Smart Tap Network
A Study on Privacy Level in Publishing Data of Smart Tap NetworkHa Phuong
 
ICIS - Power price prediction with neural networks
ICIS - Power price prediction with neural networksICIS - Power price prediction with neural networks
ICIS - Power price prediction with neural networksICIS
 
Deep learning notes.pptx
Deep learning notes.pptxDeep learning notes.pptx
Deep learning notes.pptxPandi Gingee
 
Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System
 Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System
Intrusion Detection in A Smart Forest-Fire Early Warning Sensory SystemRatul Alahy
 
AI Class Topic 6: Easy Way to Learn Deep Learning AI Technologies
AI Class Topic 6: Easy Way to Learn Deep Learning AI TechnologiesAI Class Topic 6: Easy Way to Learn Deep Learning AI Technologies
AI Class Topic 6: Easy Way to Learn Deep Learning AI TechnologiesValue Amplify Consulting
 
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...cscpconf
 
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...csandit
 
Biomedical Signals Classification With Transformer Based Model.pptx
Biomedical Signals Classification With Transformer Based Model.pptxBiomedical Signals Classification With Transformer Based Model.pptx
Biomedical Signals Classification With Transformer Based Model.pptxSandeep Kumar
 
NeuralProcessingofGeneralPurposeApproximatePrograms
NeuralProcessingofGeneralPurposeApproximateProgramsNeuralProcessingofGeneralPurposeApproximatePrograms
NeuralProcessingofGeneralPurposeApproximateProgramsMohid Nabil
 
Towards neuralprocessingofgeneralpurposeapproximateprograms
Towards neuralprocessingofgeneralpurposeapproximateprogramsTowards neuralprocessingofgeneralpurposeapproximateprograms
Towards neuralprocessingofgeneralpurposeapproximateprogramsParidha Saxena
 
Black-box Behavioral Model Inference for Autopilot Software Systems
Black-box Behavioral Model Inference for Autopilot Software SystemsBlack-box Behavioral Model Inference for Autopilot Software Systems
Black-box Behavioral Model Inference for Autopilot Software SystemsMohammad Jafar Mashhadi
 
From sensor readings to prediction: on the process of developing practical so...
From sensor readings to prediction: on the process of developing practical so...From sensor readings to prediction: on the process of developing practical so...
From sensor readings to prediction: on the process of developing practical so...Manuel Martín
 
Artificial Neural Networks for Storm Surge Prediction in North Carolina
Artificial Neural Networks for Storm Surge Prediction in North CarolinaArtificial Neural Networks for Storm Surge Prediction in North Carolina
Artificial Neural Networks for Storm Surge Prediction in North CarolinaAnton Bezuglov
 
Lab 6 Neural Network
Lab 6 Neural NetworkLab 6 Neural Network
Lab 6 Neural NetworkKyle Villano
 
PhilipSamDavisResume
PhilipSamDavisResumePhilipSamDavisResume
PhilipSamDavisResumePhilip Davis
 

Semelhante a Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks (20)

Ann model and its application
Ann model and its applicationAnn model and its application
Ann model and its application
 
A Study on Privacy Level in Publishing Data of Smart Tap Network
A Study on Privacy Level in Publishing Data of Smart Tap NetworkA Study on Privacy Level in Publishing Data of Smart Tap Network
A Study on Privacy Level in Publishing Data of Smart Tap Network
 
ICIS - Power price prediction with neural networks
ICIS - Power price prediction with neural networksICIS - Power price prediction with neural networks
ICIS - Power price prediction with neural networks
 
Deep learning notes.pptx
Deep learning notes.pptxDeep learning notes.pptx
Deep learning notes.pptx
 
Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System
 Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System
Intrusion Detection in A Smart Forest-Fire Early Warning Sensory System
 
AI Class Topic 6: Easy Way to Learn Deep Learning AI Technologies
AI Class Topic 6: Easy Way to Learn Deep Learning AI TechnologiesAI Class Topic 6: Easy Way to Learn Deep Learning AI Technologies
AI Class Topic 6: Easy Way to Learn Deep Learning AI Technologies
 
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...
Neural Networks for High Performance Time-Delay Estimation and Acoustic Sourc...
 
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...
NEURAL NETWORKS FOR HIGH PERFORMANCE TIME-DELAY ESTIMATION AND ACOUSTIC SOURC...
 
Deep Learning for Fast Simulation
Deep Learning for Fast SimulationDeep Learning for Fast Simulation
Deep Learning for Fast Simulation
 
D0521522
D0521522D0521522
D0521522
 
Biomedical Signals Classification With Transformer Based Model.pptx
Biomedical Signals Classification With Transformer Based Model.pptxBiomedical Signals Classification With Transformer Based Model.pptx
Biomedical Signals Classification With Transformer Based Model.pptx
 
NeuralProcessingofGeneralPurposeApproximatePrograms
NeuralProcessingofGeneralPurposeApproximateProgramsNeuralProcessingofGeneralPurposeApproximatePrograms
NeuralProcessingofGeneralPurposeApproximatePrograms
 
Towards neuralprocessingofgeneralpurposeapproximateprograms
Towards neuralprocessingofgeneralpurposeapproximateprogramsTowards neuralprocessingofgeneralpurposeapproximateprograms
Towards neuralprocessingofgeneralpurposeapproximateprograms
 
Machine Learning
Machine LearningMachine Learning
Machine Learning
 
Black-box Behavioral Model Inference for Autopilot Software Systems
Black-box Behavioral Model Inference for Autopilot Software SystemsBlack-box Behavioral Model Inference for Autopilot Software Systems
Black-box Behavioral Model Inference for Autopilot Software Systems
 
From sensor readings to prediction: on the process of developing practical so...
From sensor readings to prediction: on the process of developing practical so...From sensor readings to prediction: on the process of developing practical so...
From sensor readings to prediction: on the process of developing practical so...
 
Artificial Neural Networks for Storm Surge Prediction in North Carolina
Artificial Neural Networks for Storm Surge Prediction in North CarolinaArtificial Neural Networks for Storm Surge Prediction in North Carolina
Artificial Neural Networks for Storm Surge Prediction in North Carolina
 
APPLICATIONS OF WAVELET TRANSFORMS AND NEURAL NETWORKS IN EARTHQUAKE GEOTECHN...
APPLICATIONS OF WAVELET TRANSFORMS AND NEURAL NETWORKS IN EARTHQUAKE GEOTECHN...APPLICATIONS OF WAVELET TRANSFORMS AND NEURAL NETWORKS IN EARTHQUAKE GEOTECHN...
APPLICATIONS OF WAVELET TRANSFORMS AND NEURAL NETWORKS IN EARTHQUAKE GEOTECHN...
 
Lab 6 Neural Network
Lab 6 Neural NetworkLab 6 Neural Network
Lab 6 Neural Network
 
PhilipSamDavisResume
PhilipSamDavisResumePhilipSamDavisResume
PhilipSamDavisResume
 

Mais de Francisco Zamora-Martinez

Integration of Unsupervised and Supervised Criteria for DNNs Training
Integration of Unsupervised and Supervised Criteria for DNNs TrainingIntegration of Unsupervised and Supervised Criteria for DNNs Training
Integration of Unsupervised and Supervised Criteria for DNNs TrainingFrancisco Zamora-Martinez
 
F-Measure as the error function to train Neural Networks
F-Measure as the error function to train Neural NetworksF-Measure as the error function to train Neural Networks
F-Measure as the error function to train Neural NetworksFrancisco Zamora-Martinez
 
Contributions to connectionist language modeling and its application to seque...
Contributions to connectionist language modeling and its application to seque...Contributions to connectionist language modeling and its application to seque...
Contributions to connectionist language modeling and its application to seque...Francisco Zamora-Martinez
 
A Connectionist approach to Part-Of-Speech Tagging
A Connectionist approach to Part-Of-Speech TaggingA Connectionist approach to Part-Of-Speech Tagging
A Connectionist approach to Part-Of-Speech TaggingFrancisco Zamora-Martinez
 
Adding morphological information to a connectionist Part-Of-Speech tagger
Adding morphological information  to a connectionist Part-Of-Speech taggerAdding morphological information  to a connectionist Part-Of-Speech tagger
Adding morphological information to a connectionist Part-Of-Speech taggerFrancisco Zamora-Martinez
 
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...Francisco Zamora-Martinez
 
Behaviour-based Clustering of Neural Networks applied to Document Enhancement
Behaviour-based Clustering of Neural Networks applied to Document EnhancementBehaviour-based Clustering of Neural Networks applied to Document Enhancement
Behaviour-based Clustering of Neural Networks applied to Document EnhancementFrancisco Zamora-Martinez
 
Efficient Viterbi algorithms for lexical tree based models
Efficient Viterbi algorithms for lexical tree based modelsEfficient Viterbi algorithms for lexical tree based models
Efficient Viterbi algorithms for lexical tree based modelsFrancisco Zamora-Martinez
 
Efficient BP Algorithms for General Feedforward Neural Networks
Efficient BP Algorithms for General Feedforward Neural NetworksEfficient BP Algorithms for General Feedforward Neural Networks
Efficient BP Algorithms for General Feedforward Neural NetworksFrancisco Zamora-Martinez
 
Fast evaluation of Connectionist Language Models
Fast evaluation of Connectionist Language ModelsFast evaluation of Connectionist Language Models
Fast evaluation of Connectionist Language ModelsFrancisco Zamora-Martinez
 
Some empirical evaluations of a temperature forecasting module based on Art...
Some empirical evaluations of a temperature forecasting module   based on Art...Some empirical evaluations of a temperature forecasting module   based on Art...
Some empirical evaluations of a temperature forecasting module based on Art...Francisco Zamora-Martinez
 

Mais de Francisco Zamora-Martinez (11)

Integration of Unsupervised and Supervised Criteria for DNNs Training
Integration of Unsupervised and Supervised Criteria for DNNs TrainingIntegration of Unsupervised and Supervised Criteria for DNNs Training
Integration of Unsupervised and Supervised Criteria for DNNs Training
 
F-Measure as the error function to train Neural Networks
F-Measure as the error function to train Neural NetworksF-Measure as the error function to train Neural Networks
F-Measure as the error function to train Neural Networks
 
Contributions to connectionist language modeling and its application to seque...
Contributions to connectionist language modeling and its application to seque...Contributions to connectionist language modeling and its application to seque...
Contributions to connectionist language modeling and its application to seque...
 
A Connectionist approach to Part-Of-Speech Tagging
A Connectionist approach to Part-Of-Speech TaggingA Connectionist approach to Part-Of-Speech Tagging
A Connectionist approach to Part-Of-Speech Tagging
 
Adding morphological information to a connectionist Part-Of-Speech tagger
Adding morphological information  to a connectionist Part-Of-Speech taggerAdding morphological information  to a connectionist Part-Of-Speech tagger
Adding morphological information to a connectionist Part-Of-Speech tagger
 
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...
Mejora del reconocimiento de palabras manuscritas aisladas mediante un clasif...
 
Behaviour-based Clustering of Neural Networks applied to Document Enhancement
Behaviour-based Clustering of Neural Networks applied to Document EnhancementBehaviour-based Clustering of Neural Networks applied to Document Enhancement
Behaviour-based Clustering of Neural Networks applied to Document Enhancement
 
Efficient Viterbi algorithms for lexical tree based models
Efficient Viterbi algorithms for lexical tree based modelsEfficient Viterbi algorithms for lexical tree based models
Efficient Viterbi algorithms for lexical tree based models
 
Efficient BP Algorithms for General Feedforward Neural Networks
Efficient BP Algorithms for General Feedforward Neural NetworksEfficient BP Algorithms for General Feedforward Neural Networks
Efficient BP Algorithms for General Feedforward Neural Networks
 
Fast evaluation of Connectionist Language Models
Fast evaluation of Connectionist Language ModelsFast evaluation of Connectionist Language Models
Fast evaluation of Connectionist Language Models
 
Some empirical evaluations of a temperature forecasting module based on Art...
Some empirical evaluations of a temperature forecasting module   based on Art...Some empirical evaluations of a temperature forecasting module   based on Art...
Some empirical evaluations of a temperature forecasting module based on Art...
 

Último

Karra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxKarra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxAshokKarra1
 
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...Nguyen Thanh Tu Collection
 
Influencing policy (training slides from Fast Track Impact)
Influencing policy (training slides from Fast Track Impact)Influencing policy (training slides from Fast Track Impact)
Influencing policy (training slides from Fast Track Impact)Mark Reed
 
Transaction Management in Database Management System
Transaction Management in Database Management SystemTransaction Management in Database Management System
Transaction Management in Database Management SystemChristalin Nelson
 
Judging the Relevance and worth of ideas part 2.pptx
Judging the Relevance  and worth of ideas part 2.pptxJudging the Relevance  and worth of ideas part 2.pptx
Judging the Relevance and worth of ideas part 2.pptxSherlyMaeNeri
 
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxiammrhaywood
 
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...Postal Advocate Inc.
 
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdf
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdfVirtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdf
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdfErwinPantujan2
 
Proudly South Africa powerpoint Thorisha.pptx
Proudly South Africa powerpoint Thorisha.pptxProudly South Africa powerpoint Thorisha.pptx
Proudly South Africa powerpoint Thorisha.pptxthorishapillay1
 
What is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPWhat is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPCeline George
 
Global Lehigh Strategic Initiatives (without descriptions)
Global Lehigh Strategic Initiatives (without descriptions)Global Lehigh Strategic Initiatives (without descriptions)
Global Lehigh Strategic Initiatives (without descriptions)cama23
 
Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Celine George
 
4.16.24 21st Century Movements for Black Lives.pptx
4.16.24 21st Century Movements for Black Lives.pptx4.16.24 21st Century Movements for Black Lives.pptx
4.16.24 21st Century Movements for Black Lives.pptxmary850239
 
4.18.24 Movement Legacies, Reflection, and Review.pptx
4.18.24 Movement Legacies, Reflection, and Review.pptx4.18.24 Movement Legacies, Reflection, and Review.pptx
4.18.24 Movement Legacies, Reflection, and Review.pptxmary850239
 
Keynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designKeynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designMIPLM
 
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfLike-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfMr Bounab Samir
 
How to Add Barcode on PDF Report in Odoo 17
How to Add Barcode on PDF Report in Odoo 17How to Add Barcode on PDF Report in Odoo 17
How to Add Barcode on PDF Report in Odoo 17Celine George
 
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)lakshayb543
 

Último (20)

Karra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxKarra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptx
 
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...
HỌC TỐT TIẾNG ANH 11 THEO CHƯƠNG TRÌNH GLOBAL SUCCESS ĐÁP ÁN CHI TIẾT - CẢ NĂ...
 
Influencing policy (training slides from Fast Track Impact)
Influencing policy (training slides from Fast Track Impact)Influencing policy (training slides from Fast Track Impact)
Influencing policy (training slides from Fast Track Impact)
 
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
 
Transaction Management in Database Management System
Transaction Management in Database Management SystemTransaction Management in Database Management System
Transaction Management in Database Management System
 
Judging the Relevance and worth of ideas part 2.pptx
Judging the Relevance  and worth of ideas part 2.pptxJudging the Relevance  and worth of ideas part 2.pptx
Judging the Relevance and worth of ideas part 2.pptx
 
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
 
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...
USPS® Forced Meter Migration - How to Know if Your Postage Meter Will Soon be...
 
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdf
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdfVirtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdf
Virtual-Orientation-on-the-Administration-of-NATG12-NATG6-and-ELLNA.pdf
 
Raw materials used in Herbal Cosmetics.pptx
Raw materials used in Herbal Cosmetics.pptxRaw materials used in Herbal Cosmetics.pptx
Raw materials used in Herbal Cosmetics.pptx
 
Proudly South Africa powerpoint Thorisha.pptx
Proudly South Africa powerpoint Thorisha.pptxProudly South Africa powerpoint Thorisha.pptx
Proudly South Africa powerpoint Thorisha.pptx
 
What is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPWhat is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERP
 
Global Lehigh Strategic Initiatives (without descriptions)
Global Lehigh Strategic Initiatives (without descriptions)Global Lehigh Strategic Initiatives (without descriptions)
Global Lehigh Strategic Initiatives (without descriptions)
 
Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17
 
4.16.24 21st Century Movements for Black Lives.pptx
4.16.24 21st Century Movements for Black Lives.pptx4.16.24 21st Century Movements for Black Lives.pptx
4.16.24 21st Century Movements for Black Lives.pptx
 
4.18.24 Movement Legacies, Reflection, and Review.pptx
4.18.24 Movement Legacies, Reflection, and Review.pptx4.18.24 Movement Legacies, Reflection, and Review.pptx
4.18.24 Movement Legacies, Reflection, and Review.pptx
 
Keynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designKeynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-design
 
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfLike-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
 
How to Add Barcode on PDF Report in Odoo 17
How to Add Barcode on PDF Report in Odoo 17How to Add Barcode on PDF Report in Odoo 17
How to Add Barcode on PDF Report in Odoo 17
 
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
 

Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks

  • 1. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks P. Romeu, F. Zamora-Mart´ınez, P. Botella-Rocamora, J. Pardo Embedded Systems and Artificial Intelligence group Departamento de ciencias f´ısicas, matem´aticas y de la computaci´on Escuela Superior de Ense˜nanzas T´ecnicas (ESET) Universidad CEU Cardenal Herrera, 46115 Alfara del Patriarca, Valencia (Spain) ICANN – September 11, 2013
  • 2. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 3. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Introduction and motivation Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 4. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Introduction and motivation Introduction and motivation Time series forecasting: prediction future values given past data. ¯s = s0,...,si−1,si,si+1,... Non-linear relationships could be found between the elements. ANNs were widely used for this task, normally shallow models. Deep architectures has been successful in computer vision, speech signal processing, classification, . . . Time series forecasting with deep architectures is starting to receive interest (as far as we know, using Restricted Boltzmann Machines).
  • 5. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Introduction and motivation Deep architectures on time series Expectations Time series are characterized by more or less complex dependencies. For indoor temperature forecasting: Known dependencies: time of the day, day of the year. Hidden dependencies: number of people in a room. Short-term dependencies and long-term dependencies. Normally, expert knowledge is introduced to take into account known dependencies; data preprocessing: detrend, deseasoned. A deep model could learn some of these dependencies using several layers.
  • 6. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Introduction and motivation Forecasting of indoor temperature with deep ANNs What have we done in this work? Evaluation of pre-training and denoising techniques in a time series forecasting task. Results: slightly better generalization, less over-fitting. Problems: lack of data, not complex enough task. 15 16 17 18 19 20 21 22 23 24 25 26 0 2000 4000 6000 8000 10000 ºC Time (minutes)
  • 7. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Stacked Denoising Auto-Encoders Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 8. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Stacked Denoising Auto-Encoders Stacked Denoising Auto-Encoders A Denoising Auto-Encoder is a neural network which receives a noisy input and produces its cleaned version. Gaussian additive noise (σ): ˙x = x+N (0,σ2 I) Masking noise (p): ˜x = MN(˙x) with p probability. Encoding: h(˜x) = so ftsign(b+W ˜x) Decoding (denoising): ˆx = g(h(˜x)) = so ftsign(c+WT h(˜x)) ˙x h(˜x) ˜x ˆx W WT x GN(x) MN(˙x) x is an input vector, h(·) is the hidden layer vector, b and c are bias vectors, W is a weights matrix, softsign(·) = x 1+|x|
  • 9. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Stacked Denoising Auto-Encoders Stacked Denoising Auto-Encoders Greedy training building layer-by-layer auto-encoders. Stack all the trained weights to produce the final result. Stack a forecasting layer (linear activation). Train the whole neural network.
  • 10. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Time series forecasting Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 11. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Time series forecasting Time series forecasting Univariate vs multivariate. Single-step-ahead vs multi-step-ahead. Iterative forecasting vs direct forecasting. Multiple Input One Output vs Multiple Input Multiple Output. ˆst+H t+1 = F(st t−I+1) MIMO modelling is natural in ANNs, because they take profit of the input/output mapping. F is a forecasting model, H the number of predicted samples, I the number of past samples taken as input.
  • 12. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 13. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Dataset Dataset Captured during 2011, March and June. 1 minute sampling period. Reduced and smoothed by computing mean every 15 samples. Differences between adjacent samples were computed to remove the trend. Partition # of samples # of days Training 2016 21 Validation 672 7 Test 672 7
  • 14. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Evaluation measures Evaluation measures Mean Absolute Error (MAE) Root Mean Square Error (RMSE) MAE (t) = 1 |D| |D| ∑ t=I 1 H H ∑ h=1 |ˆst+h −st+h| RMSE (t) = 1 |D| |D| ∑ t=I 1 H H ∑ h=1 (ˆst+h −st+h)2 |D| is the size of the dataset, H the future horizon, ˆst+h the forecasted value, st+h the ground truth.
  • 15. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Experiments Experiments Different training modes comparison TM-0 consists in a standard training of an ANN. TM-1 pre-train the ANN using SDAE and fine-tuning of the whole network TM-2 pre-train the ANN using SDAE and fine-tuning of only last layer (forecasting layer).
  • 16. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Experiments Experiments Training description Back-propagation with mini-batch size 32. Mean Square Error (MSE) loss function. Future horizon of 12 samples (three hours). Minimum of 50 epochs, maximum of 4000. Random search hyper-parameter optimization: learning rate, momentum, weight decay, number of hidden layers, hidden layer sizes, number of inputs, mask noise percentage. 3600 experiments for tuning.
  • 17. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Results Results Best topologies - TM-0: 60 — 756 — 60 — 12 - TM-1: 48 — 648 — 920 — 16 — 12 - TM-2: 96 — 712 — 12 TM-0 has convergence problems with deep networks: 33% of two layered network experiments do not converge. 58% of three layered network experiments do not converge. Note that the topologies are not the same in the three cases, we took the best topology for each training mode.
  • 18. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Results Results 20 random initializations of best hyper-parameters 0.115 0.120 0.125 0.130 0.135 0.140 TM-0 TM-1 TM-2 MAE* Validation Test 0.135 0.140 0.145 0.150 0.155 0.160 0.165 0.170 TM-0 TM-1 TM-2 RMSE*
  • 19. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Results Results MSE of training partition during training 0.010 0.014 0.019 0.025 0.034 0.046 0.063 0.086 0.117 0.159 0 200 400 600 800 1000 1200 1400 TrainingMSE(log-scaled) Epochs TM-0 TM-1 TM-2
  • 20. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Experimentation Results Results MAE of test partition during training 0.117 0.159 0.216 0.293 0.398 0 200 400 600 800 1000 1200 1400 TestMAE*(log-scaled) Epochs best val TM-0 best val TM-1 best val TM-2 TM-0 TM-1 TM-2
  • 21. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Conclusions and future work Index 1 Introduction and motivation 2 Stacked Denoising Auto-Encoders 3 Time series forecasting 4 Experimentation 5 Conclusions and future work
  • 22. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Conclusions and future work Conclusions and future work Pre-training, denoising techniques, and random hyper-parameter optimization were used to carry out deep ANNs training in a forecasting task. Slightly better generalization performance at test set and a reduction in over-fitting was observed (TM-1). Fine-tuning phase of the whole deep model was needed to ensure good results (TM-1 vs TM-2). The short benefit of SDAE could be due to the low dimensionality of the task. In the future, this work will be extended by using larger forecasting input window combined with multivariate forecasting.
  • 23. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Conclusions and future work Questions? Thanks for your attention!
  • 24. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: Hyper-parameter optimization Grid search part Train Mode: TM-0, TM-1, TM-2 Number of hidden layers: 1, 2, 3 Mask Noise: 0.02, 0.04, 0.10, 0.20 Random search part 100 random trials for every grid sweep Input size: 12, 24, 36, 48, 60, 72, 84, 96 Learning rate: [10−3 ,10−2 ] Momentum: ∼ N (10−3 ,5×10−3 ), ignoring negative values Weight decay: [0,10−5 ] Hidden layer sizes: [4,1024]
  • 25. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: hyper-parameters analysis Input size 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.20 0.21 12 36 60 84 TM-0 12 36 60 84 TM-1 1 layer 2 layers 3 layers 12 36 60 84 TM-2
  • 26. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: hyper-parameters analysis Encoding layer size 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.20 0.21 0 300 600 900 TM-0 0 300 600 900 TM-1 0 300 600 900 TM-2
  • 27. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: hyper-parameters analysis Masking noise 0.12 0.13 0.14 0.15 0.16 0.17 0.02 0.10 0.18 TM-0 0.02 0.10 0.18 TM-1 0.02 0.10 0.18 TM-2
  • 28. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: hyper-parameters analysis Learning rate of forecasting phase 0.12 0.13 0.14 0.15 0.16 0.17 0 0.003 0.006 0.009 TM-0 0 0.003 0.006 0.009 TM-1 0 0.003 0.006 0.009 TM-2
  • 29. Time-series forecasting of indoor temperature using pre-trained Deep Neural Networks Appendix Appendix: results table MAE Validation (µ±σ) Test (µ±σ) ETS 0.3004 0.3254 TM-0 0.1289±0.0011 0.12482±0.0010 TM-1 0.1287±0.0033 0.1223±0.0033 TM-2 0.1374±0.0007 0.1279±0.0011 RMSE Validation (µ±σ) Test (µ±σ) ETS 0.3648 0.3930 TM-0 0.1563±0.0011 0.1511±0.0012 TM-1 0.1565±0.0040 0.1473±0.0039 TM-2 0.1663±0.0009 0.1538±0.0013