SlideShare a Scribd company logo
1 of 8
Senior Researcher
Masayuki Tanaka
2018/12/27
Social Intelligence Research Team
Year-End Seminar
http://www.ok.sc.e.titech.ac.jp/~mtanaka
https://twitter.com/likesilkto
Recent Conference Presentations
Gradient-Based Low-Light Image Enhancement
Masayuki Tanaka, Takashi Shibata and Masatoshi Okutomi
ProceedingsofIEEEInternationalConferenceonConsumerElectronics(ICCE2019),January,2019
Pixelwise JPEG compression detection and quality factor
estimation based on convolutional neural network
Kazutaka Uchida, Masayuki Tanaka, and Masatoshi Okutomi
Proceedings of IS&T International Symposium on Electronic Imaging (EI2019), January, 2019
Disparity Map Estimation from Cross-Modal Stereo
Thapanapong Rukkanchanunt, Takashi Shibata, Masayuki Tanaka and Masatoshi Okutomi
Proceedings of 6th IEEE Global Conference on Signal and Information Processing
(GlobalSIP2018), pp.988-992, November, 2018
Non-blindImageRestorationBasedonConvolutional NeuralNetwork
Kazutaka Uchida, Masayuki Tanaka and Masatoshi Okutomi
ProceedingsofIEEE7thGlobalConferenceonConsumerElectronics(GCCE2018),pp.12-16,October,2018
RemoteHeartRateMeasurementfromRGB-NIRVideoBasedonSpatial
andSpectralFacePatchSelection
ShiikaKado,YusukeMonno,KentaMoriwaki,KazunoriYoshizaki,MasayukiTanakaandMasatoshiOkutomi
Proceedings of 40th International Conference of the IEEE Engineering in Medicine and Biology
Society (EMBC2018), pp.5676-5680, July, 2018
Activation Functions for DNNs
Input
x
Activation
function
Weight
Output
y
Conv.
Activation
function
Input
x
Output
y
Activation functions
Sigmoid tanh ReLU
𝜎𝜎 𝑥𝑥 =
1
1 + 𝑒𝑒−𝑥𝑥
max(𝑥𝑥, 0)
Advanced Activation Functions
ReLU
max(𝑥𝑥, 0)
�
𝑥𝑥 (𝑥𝑥 ≥ 0)
𝛼𝛼 (𝑥𝑥 < 0)
Leaky ReLU
Parametric ReLU
swish, SiL
𝑥𝑥 𝜎𝜎 𝑤𝑤𝑤𝑤 + 𝑏𝑏
Existing activation functions are
element-wise function.
Dying ReLU:
Dead ReLU units always
return zero.
WiG: Weighted Sigmoid Gate (Proposed)
Existing activation functions are
element-wise function.
Sigmoid Gated Network can be
used as activation function.
Weight
Activation
function
Weight
Activation
networkunit
Proposed WiG (Weighted sigmoid gate unit)
W ×
Wg
WiG activation unit
It is compatible to existing activation functions.
It includes the ReLU.
Sigmoid
W
Wg
×
My recommendation is:
You can improve the network performance just by
replacing the ReLU by the proposed WiG.
Experimental Validations
Object recognition
Average accuracy
Image denoising
The reproduction code is available
http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/WiG/
Reference
Masayuki Tanaka, Weighted Sigmoid Gate Unit for an Activation Function
of Deep Neural Network, arXiv preprint arXiv:1810.01829, 2018.
http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/WiG/
V. Nair and G. E. Hinton, “Rectified linear units improve restricted boltzmann
machines,” in Proceedings of the 27th international conference on machine
learning (ICML-10), 2010, pp. 807–814.
P. Ramachandran, B. Zoph, and Q. V. Le, “Searching for activation functions,” arXiv
preprint arXiv:1710.05941, 2017.
S. Elfwing, E. Uchibe, and K. Doya, “Sigmoid-weighted linear units for neural
network function approximation in reinforcement learning,”arXiv preprint
arXiv:1702.03118, 2017.
A. L. Maas, A. Y. Hannun, and A. Y. Ng, “Rectifier nonlinearities improve neural
network acoustic models,” in International Conference on Machine Learning (ICML),
vol. 30, no. 1, 2013, p. 3.
K. He, X. Zhang, S. Ren, and J. Sun, “Delving deep into rectifiers: Surpassing
human-level performance on imagenet classification,” in IEEE International
Conference on Computer Vision (ICCV), 2015, pp. 1026–1034.
Promotion: Train1000 project
The deep learning usually requires a huge size of training data to improve the
performance. However, a training with such a huge data needs high computational
cost in terms of both computational power and time. It is very tough especially for
beginners. In practice, it is hard to collect a huge number of annotated training
samples. I think that 1,000 samples are minimum number for the training of the
network. The training with 1,000 samples also includes technical challenges. One
of them is to improve generalization performance while avoiding the over fitting.
Let’s enjoy train with 1000.
http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/train1000/
Sample codes of matlab and keras for mnits and cifar are available.

More Related Content

What's hot

WANG Qian_CV_2016
WANG Qian_CV_2016WANG Qian_CV_2016
WANG Qian_CV_2016qian Wang
 
September 2021 - Top 10 Read Articles in Signal & Image Processing
September 2021 - Top 10 Read Articles in Signal & Image ProcessingSeptember 2021 - Top 10 Read Articles in Signal & Image Processing
September 2021 - Top 10 Read Articles in Signal & Image Processingsipij
 
Augmented Collective Digital Twins for Self-Organising Cyber-Physical Systems
Augmented Collective Digital Twins for Self-Organising Cyber-Physical SystemsAugmented Collective Digital Twins for Self-Organising Cyber-Physical Systems
Augmented Collective Digital Twins for Self-Organising Cyber-Physical SystemsRoberto Casadei
 
Artificial Intelligences Using Silicon and Beyond Conference
Artificial Intelligences Using Silicon and Beyond ConferenceArtificial Intelligences Using Silicon and Beyond Conference
Artificial Intelligences Using Silicon and Beyond Conferenceobject arena
 
Feasibility of Artificial Neural Network in Civil Engineering
Feasibility of Artificial Neural Network in Civil EngineeringFeasibility of Artificial Neural Network in Civil Engineering
Feasibility of Artificial Neural Network in Civil Engineeringijtsrd
 
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...Shrikant Kate
 
Artificial Intelligence in Civil Engineering.
 Artificial Intelligence in Civil Engineering.  Artificial Intelligence in Civil Engineering.
Artificial Intelligence in Civil Engineering. hannan366
 
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...semanticsconference
 

What's hot (17)

WANG Qian_CV_2016
WANG Qian_CV_2016WANG Qian_CV_2016
WANG Qian_CV_2016
 
RESUME
RESUME RESUME
RESUME
 
Vedansh thakkar resume
Vedansh thakkar resumeVedansh thakkar resume
Vedansh thakkar resume
 
Chepuri Srilaxmi
Chepuri SrilaxmiChepuri Srilaxmi
Chepuri Srilaxmi
 
September 2021 - Top 10 Read Articles in Signal & Image Processing
September 2021 - Top 10 Read Articles in Signal & Image ProcessingSeptember 2021 - Top 10 Read Articles in Signal & Image Processing
September 2021 - Top 10 Read Articles in Signal & Image Processing
 
Visible Light Communications for Li-Fi Technology Using PWM Signals
Visible Light Communications for Li-Fi Technology Using PWM SignalsVisible Light Communications for Li-Fi Technology Using PWM Signals
Visible Light Communications for Li-Fi Technology Using PWM Signals
 
Augmented Collective Digital Twins for Self-Organising Cyber-Physical Systems
Augmented Collective Digital Twins for Self-Organising Cyber-Physical SystemsAugmented Collective Digital Twins for Self-Organising Cyber-Physical Systems
Augmented Collective Digital Twins for Self-Organising Cyber-Physical Systems
 
Artificial Intelligences Using Silicon and Beyond Conference
Artificial Intelligences Using Silicon and Beyond ConferenceArtificial Intelligences Using Silicon and Beyond Conference
Artificial Intelligences Using Silicon and Beyond Conference
 
janath tele resume
janath tele resumejanath tele resume
janath tele resume
 
Feasibility of Artificial Neural Network in Civil Engineering
Feasibility of Artificial Neural Network in Civil EngineeringFeasibility of Artificial Neural Network in Civil Engineering
Feasibility of Artificial Neural Network in Civil Engineering
 
riadul_islam_cv
riadul_islam_cvriadul_islam_cv
riadul_islam_cv
 
Resume
ResumeResume
Resume
 
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...
Introduction to Artificial Intelligence Technique for Civil Engineering_ Unit...
 
Artificial Intelligence in Civil Engineering.
 Artificial Intelligence in Civil Engineering.  Artificial Intelligence in Civil Engineering.
Artificial Intelligence in Civil Engineering.
 
Academic_CV_Letter
Academic_CV_LetterAcademic_CV_Letter
Academic_CV_Letter
 
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...
Fajar J. Ekaputra, Marta Sabou, Estefania Serral and Stefan Biffl | Knowledge...
 
Niharika Resume
Niharika ResumeNiharika Resume
Niharika Resume
 

Similar to Senior Researcher's Year-End Seminar on Deep Learning Advances

New research articles 2020 october issue international journal of multimedi...
New research articles 2020 october  issue  international journal of multimedi...New research articles 2020 october  issue  international journal of multimedi...
New research articles 2020 october issue international journal of multimedi...ijma
 
November 2021: Top Read Articles in Signal & Image Processing
November 2021: Top Read Articles in Signal & Image ProcessingNovember 2021: Top Read Articles in Signal & Image Processing
November 2021: Top Read Articles in Signal & Image Processingsipij
 
December 2021: Top Read Articles in Signal & Image Processing
December 2021: Top Read Articles in Signal & Image ProcessingDecember 2021: Top Read Articles in Signal & Image Processing
December 2021: Top Read Articles in Signal & Image Processingsipij
 
July 2021: Top Read Articles in Signal & Image Processing
July 2021: Top Read Articles in Signal & Image ProcessingJuly 2021: Top Read Articles in Signal & Image Processing
July 2021: Top Read Articles in Signal & Image Processingsipij
 
June 2021: Top Read Articles in Signal & Image Processing
June 2021: Top Read Articles in Signal & Image ProcessingJune 2021: Top Read Articles in Signal & Image Processing
June 2021: Top Read Articles in Signal & Image Processingsipij
 
Jia-Bin Huang's Curriculum Vitae
Jia-Bin Huang's Curriculum VitaeJia-Bin Huang's Curriculum Vitae
Jia-Bin Huang's Curriculum VitaeJia-Bin Huang
 
CV_Lisongnan2015
CV_Lisongnan2015CV_Lisongnan2015
CV_Lisongnan2015Songnan Li
 
June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing  June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing ijsc
 
TOP 5 Most View Article From Academia in 2019
TOP 5 Most View Article From Academia in 2019TOP 5 Most View Article From Academia in 2019
TOP 5 Most View Article From Academia in 2019sipij
 
Recent articles published in VLSI design & Communication Systems
 Recent articles published in VLSI design & Communication Systems Recent articles published in VLSI design & Communication Systems
Recent articles published in VLSI design & Communication SystemsVLSICS Design
 
New Research Articles 2019 September Issue International Journal of Artificia...
New Research Articles 2019 September Issue International Journal of Artificia...New Research Articles 2019 September Issue International Journal of Artificia...
New Research Articles 2019 September Issue International Journal of Artificia...gerogepatton
 
Top Cited Article in Informatics Engineering Research: October 2020
Top Cited Article in Informatics Engineering Research: October 2020Top Cited Article in Informatics Engineering Research: October 2020
Top Cited Article in Informatics Engineering Research: October 2020ieijjournal
 
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfFederated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfKundjanasith Thonglek
 
CVLinkedIn
CVLinkedInCVLinkedIn
CVLinkedInJun Ma
 
CV-English.doc
CV-English.docCV-English.doc
CV-English.docbutest
 
CV-English.doc
CV-English.docCV-English.doc
CV-English.docbutest
 
TOP READ NATURAL LANGUAGE COMPUTING ARTICLE 2020
TOP READ NATURAL LANGUAGE  COMPUTING ARTICLE 2020TOP READ NATURAL LANGUAGE  COMPUTING ARTICLE 2020
TOP READ NATURAL LANGUAGE COMPUTING ARTICLE 2020kevig
 
Top 20 Cited Article in Computer Science & Information Technology
Top 20 Cited Article in Computer Science & Information TechnologyTop 20 Cited Article in Computer Science & Information Technology
Top 20 Cited Article in Computer Science & Information TechnologyAIRCC Publishing Corporation
 
Deep randomized neural networks
Deep randomized neural networksDeep randomized neural networks
Deep randomized neural networksClaudio Gallicchio
 

Similar to Senior Researcher's Year-End Seminar on Deep Learning Advances (20)

New research articles 2020 october issue international journal of multimedi...
New research articles 2020 october  issue  international journal of multimedi...New research articles 2020 october  issue  international journal of multimedi...
New research articles 2020 october issue international journal of multimedi...
 
November 2021: Top Read Articles in Signal & Image Processing
November 2021: Top Read Articles in Signal & Image ProcessingNovember 2021: Top Read Articles in Signal & Image Processing
November 2021: Top Read Articles in Signal & Image Processing
 
December 2021: Top Read Articles in Signal & Image Processing
December 2021: Top Read Articles in Signal & Image ProcessingDecember 2021: Top Read Articles in Signal & Image Processing
December 2021: Top Read Articles in Signal & Image Processing
 
July 2021: Top Read Articles in Signal & Image Processing
July 2021: Top Read Articles in Signal & Image ProcessingJuly 2021: Top Read Articles in Signal & Image Processing
July 2021: Top Read Articles in Signal & Image Processing
 
June 2021: Top Read Articles in Signal & Image Processing
June 2021: Top Read Articles in Signal & Image ProcessingJune 2021: Top Read Articles in Signal & Image Processing
June 2021: Top Read Articles in Signal & Image Processing
 
Jia-Bin Huang's Curriculum Vitae
Jia-Bin Huang's Curriculum VitaeJia-Bin Huang's Curriculum Vitae
Jia-Bin Huang's Curriculum Vitae
 
CV_Lisongnan2015
CV_Lisongnan2015CV_Lisongnan2015
CV_Lisongnan2015
 
June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing  June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing
 
TOP 5 Most View Article From Academia in 2019
TOP 5 Most View Article From Academia in 2019TOP 5 Most View Article From Academia in 2019
TOP 5 Most View Article From Academia in 2019
 
Recent articles published in VLSI design & Communication Systems
 Recent articles published in VLSI design & Communication Systems Recent articles published in VLSI design & Communication Systems
Recent articles published in VLSI design & Communication Systems
 
New Research Articles 2019 September Issue International Journal of Artificia...
New Research Articles 2019 September Issue International Journal of Artificia...New Research Articles 2019 September Issue International Journal of Artificia...
New Research Articles 2019 September Issue International Journal of Artificia...
 
Top Cited Article in Informatics Engineering Research: October 2020
Top Cited Article in Informatics Engineering Research: October 2020Top Cited Article in Informatics Engineering Research: October 2020
Top Cited Article in Informatics Engineering Research: October 2020
 
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfFederated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
 
CVLinkedIn
CVLinkedInCVLinkedIn
CVLinkedIn
 
CV-English.doc
CV-English.docCV-English.doc
CV-English.doc
 
CV-English.doc
CV-English.docCV-English.doc
CV-English.doc
 
TOP READ NATURAL LANGUAGE COMPUTING ARTICLE 2020
TOP READ NATURAL LANGUAGE  COMPUTING ARTICLE 2020TOP READ NATURAL LANGUAGE  COMPUTING ARTICLE 2020
TOP READ NATURAL LANGUAGE COMPUTING ARTICLE 2020
 
Top 20 Cited Article in Computer Science & Information Technology
Top 20 Cited Article in Computer Science & Information TechnologyTop 20 Cited Article in Computer Science & Information Technology
Top 20 Cited Article in Computer Science & Information Technology
 
Deep randomized neural networks
Deep randomized neural networksDeep randomized neural networks
Deep randomized neural networks
 
Learning with Unpaired Data
Learning with Unpaired DataLearning with Unpaired Data
Learning with Unpaired Data
 

More from Masayuki Tanaka

Slideshare breaking inter layer co-adaptation
Slideshare breaking inter layer co-adaptationSlideshare breaking inter layer co-adaptation
Slideshare breaking inter layer co-adaptationMasayuki Tanaka
 
PRMU201902 Presentation document
PRMU201902 Presentation documentPRMU201902 Presentation document
PRMU201902 Presentation documentMasayuki Tanaka
 
Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementMasayuki Tanaka
 
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得遠赤外線カメラと可視カメラを利用した悪条件下における画像取得
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得Masayuki Tanaka
 
Learnable Image Encryption
Learnable Image EncryptionLearnable Image Encryption
Learnable Image EncryptionMasayuki Tanaka
 
クリエイティブ・コモンズ
クリエイティブ・コモンズクリエイティブ・コモンズ
クリエイティブ・コモンズMasayuki Tanaka
 
メラビアンの法則
メラビアンの法則メラビアンの法則
メラビアンの法則Masayuki Tanaka
 
権威に訴える論証
権威に訴える論証権威に訴える論証
権威に訴える論証Masayuki Tanaka
 
Chain rule of deep neural network layer for back propagation
Chain rule of deep neural network layer for back propagationChain rule of deep neural network layer for back propagation
Chain rule of deep neural network layer for back propagationMasayuki Tanaka
 
One-point for presentation
One-point for presentationOne-point for presentation
One-point for presentationMasayuki Tanaka
 
ADMM algorithm in ProxImaL
ADMM algorithm in ProxImaL ADMM algorithm in ProxImaL
ADMM algorithm in ProxImaL Masayuki Tanaka
 
Intensity Constraint Gradient-Based Image Reconstruction
Intensity Constraint Gradient-Based Image ReconstructionIntensity Constraint Gradient-Based Image Reconstruction
Intensity Constraint Gradient-Based Image ReconstructionMasayuki Tanaka
 
Least Square with L0, L1, and L2 Constraint
Least Square with L0, L1, and L2 ConstraintLeast Square with L0, L1, and L2 Constraint
Least Square with L0, L1, and L2 ConstraintMasayuki Tanaka
 

More from Masayuki Tanaka (20)

Slideshare breaking inter layer co-adaptation
Slideshare breaking inter layer co-adaptationSlideshare breaking inter layer co-adaptation
Slideshare breaking inter layer co-adaptation
 
PRMU201902 Presentation document
PRMU201902 Presentation documentPRMU201902 Presentation document
PRMU201902 Presentation document
 
Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image Enhancement
 
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得遠赤外線カメラと可視カメラを利用した悪条件下における画像取得
遠赤外線カメラと可視カメラを利用した悪条件下における画像取得
 
Learnable Image Encryption
Learnable Image EncryptionLearnable Image Encryption
Learnable Image Encryption
 
クリエイティブ・コモンズ
クリエイティブ・コモンズクリエイティブ・コモンズ
クリエイティブ・コモンズ
 
デザイン4原則
デザイン4原則デザイン4原則
デザイン4原則
 
メラビアンの法則
メラビアンの法則メラビアンの法則
メラビアンの法則
 
類似性の法則
類似性の法則類似性の法則
類似性の法則
 
権威に訴える論証
権威に訴える論証権威に訴える論証
権威に訴える論証
 
Chain rule of deep neural network layer for back propagation
Chain rule of deep neural network layer for back propagationChain rule of deep neural network layer for back propagation
Chain rule of deep neural network layer for back propagation
 
Give Me Four
Give Me FourGive Me Four
Give Me Four
 
Tech art 20170315
Tech art 20170315Tech art 20170315
Tech art 20170315
 
My Slide Theme
My Slide ThemeMy Slide Theme
My Slide Theme
 
Font Memo
Font MemoFont Memo
Font Memo
 
One-point for presentation
One-point for presentationOne-point for presentation
One-point for presentation
 
ADMM algorithm in ProxImaL
ADMM algorithm in ProxImaL ADMM algorithm in ProxImaL
ADMM algorithm in ProxImaL
 
Intensity Constraint Gradient-Based Image Reconstruction
Intensity Constraint Gradient-Based Image ReconstructionIntensity Constraint Gradient-Based Image Reconstruction
Intensity Constraint Gradient-Based Image Reconstruction
 
Least Square with L0, L1, and L2 Constraint
Least Square with L0, L1, and L2 ConstraintLeast Square with L0, L1, and L2 Constraint
Least Square with L0, L1, and L2 Constraint
 
Lasso regression
Lasso regressionLasso regression
Lasso regression
 

Recently uploaded

linear Regression, multiple Regression and Annova
linear Regression, multiple Regression and Annovalinear Regression, multiple Regression and Annova
linear Regression, multiple Regression and AnnovaMansi Rastogi
 
Advances in AI-driven Image Recognition for Early Detection of Cancer
Advances in AI-driven Image Recognition for Early Detection of CancerAdvances in AI-driven Image Recognition for Early Detection of Cancer
Advances in AI-driven Image Recognition for Early Detection of CancerLuis Miguel Chong Chong
 
Q4-Mod-1c-Quiz-Projectile-333344444.pptx
Q4-Mod-1c-Quiz-Projectile-333344444.pptxQ4-Mod-1c-Quiz-Projectile-333344444.pptx
Q4-Mod-1c-Quiz-Projectile-333344444.pptxtuking87
 
DNA isolation molecular biology practical.pptx
DNA isolation molecular biology practical.pptxDNA isolation molecular biology practical.pptx
DNA isolation molecular biology practical.pptxGiDMOh
 
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer Zahana
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer ZahanaEGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer Zahana
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer ZahanaDr.Mahmoud Abbas
 
FBI Profiling - Forensic Psychology.pptx
FBI Profiling - Forensic Psychology.pptxFBI Profiling - Forensic Psychology.pptx
FBI Profiling - Forensic Psychology.pptxPayal Shrivastava
 
final waves properties grade 7 - third quarter
final waves properties grade 7 - third quarterfinal waves properties grade 7 - third quarter
final waves properties grade 7 - third quarterHanHyoKim
 
whole genome sequencing new and its types including shortgun and clone by clone
whole genome sequencing new  and its types including shortgun and clone by clonewhole genome sequencing new  and its types including shortgun and clone by clone
whole genome sequencing new and its types including shortgun and clone by clonechaudhary charan shingh university
 
Introduction of Human Body & Structure of cell.pptx
Introduction of Human Body & Structure of cell.pptxIntroduction of Human Body & Structure of cell.pptx
Introduction of Human Body & Structure of cell.pptxMedical College
 
Measures of Central Tendency.pptx for UG
Measures of Central Tendency.pptx for UGMeasures of Central Tendency.pptx for UG
Measures of Central Tendency.pptx for UGSoniaBajaj10
 
Unveiling the Cannabis Plant’s Potential
Unveiling the Cannabis Plant’s PotentialUnveiling the Cannabis Plant’s Potential
Unveiling the Cannabis Plant’s PotentialMarkus Roggen
 
BACTERIAL SECRETION SYSTEM by Dr. Chayanika Das
BACTERIAL SECRETION SYSTEM by Dr. Chayanika DasBACTERIAL SECRETION SYSTEM by Dr. Chayanika Das
BACTERIAL SECRETION SYSTEM by Dr. Chayanika DasChayanika Das
 
Abnormal LFTs rate of deco and NAFLD.pptx
Abnormal LFTs rate of deco and NAFLD.pptxAbnormal LFTs rate of deco and NAFLD.pptx
Abnormal LFTs rate of deco and NAFLD.pptxzeus70441
 
Probability.pptx, Types of Probability, UG
Probability.pptx, Types of Probability, UGProbability.pptx, Types of Probability, UG
Probability.pptx, Types of Probability, UGSoniaBajaj10
 
Loudspeaker- direct radiating type and horn type.pptx
Loudspeaker- direct radiating type and horn type.pptxLoudspeaker- direct radiating type and horn type.pptx
Loudspeaker- direct radiating type and horn type.pptxpriyankatabhane
 
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdf
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdfKDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdf
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdfGABYFIORELAMALPARTID1
 
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11GelineAvendao
 

Recently uploaded (20)

Interferons.pptx.
Interferons.pptx.Interferons.pptx.
Interferons.pptx.
 
linear Regression, multiple Regression and Annova
linear Regression, multiple Regression and Annovalinear Regression, multiple Regression and Annova
linear Regression, multiple Regression and Annova
 
Advances in AI-driven Image Recognition for Early Detection of Cancer
Advances in AI-driven Image Recognition for Early Detection of CancerAdvances in AI-driven Image Recognition for Early Detection of Cancer
Advances in AI-driven Image Recognition for Early Detection of Cancer
 
Q4-Mod-1c-Quiz-Projectile-333344444.pptx
Q4-Mod-1c-Quiz-Projectile-333344444.pptxQ4-Mod-1c-Quiz-Projectile-333344444.pptx
Q4-Mod-1c-Quiz-Projectile-333344444.pptx
 
DNA isolation molecular biology practical.pptx
DNA isolation molecular biology practical.pptxDNA isolation molecular biology practical.pptx
DNA isolation molecular biology practical.pptx
 
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer Zahana
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer ZahanaEGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer Zahana
EGYPTIAN IMPRINT IN SPAIN Lecture by Dr Abeer Zahana
 
FBI Profiling - Forensic Psychology.pptx
FBI Profiling - Forensic Psychology.pptxFBI Profiling - Forensic Psychology.pptx
FBI Profiling - Forensic Psychology.pptx
 
final waves properties grade 7 - third quarter
final waves properties grade 7 - third quarterfinal waves properties grade 7 - third quarter
final waves properties grade 7 - third quarter
 
whole genome sequencing new and its types including shortgun and clone by clone
whole genome sequencing new  and its types including shortgun and clone by clonewhole genome sequencing new  and its types including shortgun and clone by clone
whole genome sequencing new and its types including shortgun and clone by clone
 
Introduction of Human Body & Structure of cell.pptx
Introduction of Human Body & Structure of cell.pptxIntroduction of Human Body & Structure of cell.pptx
Introduction of Human Body & Structure of cell.pptx
 
Measures of Central Tendency.pptx for UG
Measures of Central Tendency.pptx for UGMeasures of Central Tendency.pptx for UG
Measures of Central Tendency.pptx for UG
 
Introduction Classification Of Alkaloids
Introduction Classification Of AlkaloidsIntroduction Classification Of Alkaloids
Introduction Classification Of Alkaloids
 
Unveiling the Cannabis Plant’s Potential
Unveiling the Cannabis Plant’s PotentialUnveiling the Cannabis Plant’s Potential
Unveiling the Cannabis Plant’s Potential
 
PLASMODIUM. PPTX
PLASMODIUM. PPTXPLASMODIUM. PPTX
PLASMODIUM. PPTX
 
BACTERIAL SECRETION SYSTEM by Dr. Chayanika Das
BACTERIAL SECRETION SYSTEM by Dr. Chayanika DasBACTERIAL SECRETION SYSTEM by Dr. Chayanika Das
BACTERIAL SECRETION SYSTEM by Dr. Chayanika Das
 
Abnormal LFTs rate of deco and NAFLD.pptx
Abnormal LFTs rate of deco and NAFLD.pptxAbnormal LFTs rate of deco and NAFLD.pptx
Abnormal LFTs rate of deco and NAFLD.pptx
 
Probability.pptx, Types of Probability, UG
Probability.pptx, Types of Probability, UGProbability.pptx, Types of Probability, UG
Probability.pptx, Types of Probability, UG
 
Loudspeaker- direct radiating type and horn type.pptx
Loudspeaker- direct radiating type and horn type.pptxLoudspeaker- direct radiating type and horn type.pptx
Loudspeaker- direct radiating type and horn type.pptx
 
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdf
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdfKDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdf
KDIGO-2023-CKD-Guideline-Public-Review-Draft_5-July-2023.pdf
 
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11
WEEK 4 PHYSICAL SCIENCE QUARTER 3 FOR G11
 

Senior Researcher's Year-End Seminar on Deep Learning Advances

  • 1. Senior Researcher Masayuki Tanaka 2018/12/27 Social Intelligence Research Team Year-End Seminar http://www.ok.sc.e.titech.ac.jp/~mtanaka https://twitter.com/likesilkto
  • 2. Recent Conference Presentations Gradient-Based Low-Light Image Enhancement Masayuki Tanaka, Takashi Shibata and Masatoshi Okutomi ProceedingsofIEEEInternationalConferenceonConsumerElectronics(ICCE2019),January,2019 Pixelwise JPEG compression detection and quality factor estimation based on convolutional neural network Kazutaka Uchida, Masayuki Tanaka, and Masatoshi Okutomi Proceedings of IS&T International Symposium on Electronic Imaging (EI2019), January, 2019 Disparity Map Estimation from Cross-Modal Stereo Thapanapong Rukkanchanunt, Takashi Shibata, Masayuki Tanaka and Masatoshi Okutomi Proceedings of 6th IEEE Global Conference on Signal and Information Processing (GlobalSIP2018), pp.988-992, November, 2018 Non-blindImageRestorationBasedonConvolutional NeuralNetwork Kazutaka Uchida, Masayuki Tanaka and Masatoshi Okutomi ProceedingsofIEEE7thGlobalConferenceonConsumerElectronics(GCCE2018),pp.12-16,October,2018 RemoteHeartRateMeasurementfromRGB-NIRVideoBasedonSpatial andSpectralFacePatchSelection ShiikaKado,YusukeMonno,KentaMoriwaki,KazunoriYoshizaki,MasayukiTanakaandMasatoshiOkutomi Proceedings of 40th International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC2018), pp.5676-5680, July, 2018
  • 3. Activation Functions for DNNs Input x Activation function Weight Output y Conv. Activation function Input x Output y Activation functions Sigmoid tanh ReLU 𝜎𝜎 𝑥𝑥 = 1 1 + 𝑒𝑒−𝑥𝑥 max(𝑥𝑥, 0)
  • 4. Advanced Activation Functions ReLU max(𝑥𝑥, 0) � 𝑥𝑥 (𝑥𝑥 ≥ 0) 𝛼𝛼 (𝑥𝑥 < 0) Leaky ReLU Parametric ReLU swish, SiL 𝑥𝑥 𝜎𝜎 𝑤𝑤𝑤𝑤 + 𝑏𝑏 Existing activation functions are element-wise function. Dying ReLU: Dead ReLU units always return zero.
  • 5. WiG: Weighted Sigmoid Gate (Proposed) Existing activation functions are element-wise function. Sigmoid Gated Network can be used as activation function. Weight Activation function Weight Activation networkunit Proposed WiG (Weighted sigmoid gate unit) W × Wg WiG activation unit It is compatible to existing activation functions. It includes the ReLU. Sigmoid W Wg × My recommendation is: You can improve the network performance just by replacing the ReLU by the proposed WiG.
  • 6. Experimental Validations Object recognition Average accuracy Image denoising The reproduction code is available http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/WiG/
  • 7. Reference Masayuki Tanaka, Weighted Sigmoid Gate Unit for an Activation Function of Deep Neural Network, arXiv preprint arXiv:1810.01829, 2018. http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/WiG/ V. Nair and G. E. Hinton, “Rectified linear units improve restricted boltzmann machines,” in Proceedings of the 27th international conference on machine learning (ICML-10), 2010, pp. 807–814. P. Ramachandran, B. Zoph, and Q. V. Le, “Searching for activation functions,” arXiv preprint arXiv:1710.05941, 2017. S. Elfwing, E. Uchibe, and K. Doya, “Sigmoid-weighted linear units for neural network function approximation in reinforcement learning,”arXiv preprint arXiv:1702.03118, 2017. A. L. Maas, A. Y. Hannun, and A. Y. Ng, “Rectifier nonlinearities improve neural network acoustic models,” in International Conference on Machine Learning (ICML), vol. 30, no. 1, 2013, p. 3. K. He, X. Zhang, S. Ren, and J. Sun, “Delving deep into rectifiers: Surpassing human-level performance on imagenet classification,” in IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1026–1034.
  • 8. Promotion: Train1000 project The deep learning usually requires a huge size of training data to improve the performance. However, a training with such a huge data needs high computational cost in terms of both computational power and time. It is very tough especially for beginners. In practice, it is hard to collect a huge number of annotated training samples. I think that 1,000 samples are minimum number for the training of the network. The training with 1,000 samples also includes technical challenges. One of them is to improve generalization performance while avoiding the over fitting. Let’s enjoy train with 1000. http://www.ok.sc.e.titech.ac.jp/~mtanaka/proj/train1000/ Sample codes of matlab and keras for mnits and cifar are available.