embc tutorial on interpretable and transparent deep
play

EMBC Tutorial on Interpretable and Transparent Deep Learning - PowerPoint PPT Presentation

Jul 11, 2023 •5 likes •545 views

EMBC Tutorial on Interpretable and Transparent Deep Learning Wojciech Samek Grgoire Montavon Klaus-Robert Mller (Fraunhofer HHI) (TU Berlin) (TU Berlin) 13:30 - 14:00 Introduction KRM 14:00 - 15:00 T echniques for Interpretability GM

  1. EMBC Tutorial on Interpretable and Transparent Deep Learning Wojciech Samek Grégoire Montavon Klaus-Robert Müller (Fraunhofer HHI) (TU Berlin) (TU Berlin) 13:30 - 14:00 Introduction KRM 14:00 - 15:00 T echniques for Interpretability GM 15:00 - 15:30 Cofgee Break ALL 15:30 - 16:15 Evaluating Interpretability & Applications WS 16:15 - 17:15 Applications in BME & the Sciences and Wrap-Up KRM 1 / 5 4

  2. N a r r o w i n g t h e C o n c e p t o f E x p l a n a t i o n 2 / 5 4

  3. E x p l a i n i n g M L M o d e l s : T w o V i e w s me c h a n i s t i c f u n c t i o n a l u n d e r s t a n d i n g u n d e r s t a n d i n g U n d e r s t a n d i n g w h a t U n d e r s t a n d i n g h o w t h e m e c h a n i s m t h e n e t w o r k n e t w o r k s r e l a t e s t h e i n p u t u s e s t o s o l v e a p r o b l e m o r t o t h e o u t p u t v a r i a b l e s . i m p l e m e n t a f u n c t i o n . 3 / 5 4

  4. E x p l a i n i n g M L M o d e l s : T w o P r o b l e ms mo d e l a n a l y s i s d e c i s i o n a n a l y s i s p o s s i b l e a p p r o a c h p o s s i b l e a p p r o a c h - b u i l d p r o t o t y p e s o f " t y p i c a l " - i d e n t i f y w h i c h i n p u t v a r i a b l e s e x a m p l e s o f a c e r t a i n c l a s s . c o n t r i b u t e t o t h e p r e d i c t i o n . 4 / 5 4

  5. E x p l a i n i n g M L M o d e l s : T w o P r o b l e ms M o d e l A n a l y s i s D e c i s i o n A n a l y s i s “ w h a t d o e s s o m e t h i n g “ w h y a g i v e n i m a g e i s c l a s s i fi e d a s p r e d i c t e d a s a p o o l t a b l e a p o o l t a b l e ” t y p i c a l l y l o o k l i k e . “ w h y i t i s c l a s s i fi e d s o m e p o o l t a b l e m o d e l ’ s p r o t o t y p i c a l a s a p o o l t a b l e p o o l t a b l e 5 / 5 4

  6. A S u r v e y o f E x p l a n a t i o n T e c h n i q u e s 6 / 5 4

  7. O v e r v i e w o f E x p l a n a t i o n M e t h o d s 1 . P e r t u r b a t i o n - B a s e d Me t h o d s 2 . Me a n i n g f u l P e r t u r b a t i o n s 3 . S i m p l e T a y l o r E x p a n s i o n 4 . G r a d i e n t x I n p u t 4 . L a y e r - Wi s e R e l e v a n c e P r o p a g a t i o n ( L R P ) 7 / 5 4

  8. A p p r o a c h 1 : P e r t u r b a t i o n Idea: Assess features relevance by testing the model response to their removal or perturbation. castle D N N D N N … still a castle not a castle D N N 8 / 5 4

  9. A p p r o a c h 1 : P e r t u r b a t i o n Building an explanation A d v a n t a g e s i=1 i=2 i=3 ... - S i m p l e . i=6 input - A p p l i c a b l e t o ML m o d e l . a n y D i s a d v a n t a g e s - N e e d t o r e e v a l u a t e t h e f u n c t i o n f o r m a n y p e r t u r b a t i o n s → s l o w i=1 i=2 i=3 ... - P e r t u r b a t i o n p r o c e s s m a y i=6 i n t r o d u c e a r t e f a c t s i n t h e i m a g e heatmap → u n r e l i a b l e 9 / 5 4

  10. A p p r o a c h 2 : M e a n i n g f u l P e r t u r b a t i o n s Idea: Don’t iterate over all possible perturbation, search locally for the best perturbation m* (or mask). Fong and Vedaldi 2017, Interpretable Explanations of Black Boxes by Meaningful Perturbation 1 0 / 5 4

  11. A p p r o a c h 2 : M e a n i n g f u l P e r t u r b a t i o n s A d v a n t a g e s - C a n b e a p p l i e d t o a n y ( d i fg e r e n t i a b l e ) ML m o d e l . L i mi t a t i o n s - N e e d t o r u n a n o p t i m i z a t i o n p r o c e d u r e 1 1 / 5 4

  12. A p p r o a c h 3 : ( S i mp l e ) T a y l o r E x p a n s i o n s I d e a : i d e n t i f y t h e c o n t r i b u t i o n o f i n p u t f e a t u r e s a s t h e fj r s t - o r d e r t e r m s o f a T a y l o r e x p a n s i o n T a y l o r E x p a n s i o n 1 2 / 5 4

  13. A p p r o a c h 3 : ( S i mp l e ) T a y l o r E x p a n s i o n s A d v a n t a g e s - C a n b e a p p l i e d t o ( d i fg e r e n t i a b l e a n d m i l d l y n o n l i n e a r ) ML m o d e l . a n y L i mi t a t i o n s - N e e d t o fj n d a m e a n i n g f u l r o o t p o i n t w h e r e t o p e r f o r m t h e e x p a n s i o n . ( → o p t i m i z a t i o n , o r h e u r i s t i c s ) 1 3 / 5 4

  14. A p p r o a c h 4 : G r a d i e n t x I n p u t Mo t i v a t i o n - C o m p u t e a n e x p l a n a t i o n i n a s i n g l e p a s s w i t h o u t h a v i n g t o o p t i m i z e o r s e a r c h f o r a r o o t p o i n t . Gradient x Input 1 4 / 5 4

  15. A p p r o a c h 4 : G r a d i e n t x I n p u t O b s e r v a t i o n : C o m p l e x a n a l y s e s r e d u c e t o g r a d i e n t x i n p u t f o r s i m p l e c a s e s . Perturbation Analysis Taylor Expansions Gradient x Input Q u e s t i o n : D o e s i t w o r k i n p r a c t i c e ? 1 5 / 5 4

  16. A p p r o a c h 4 : G r a d i e n t x I n p u t E x p l a n a t i o n I n p u t P r e d i c t i o n ( c l a s s : b a s e b a l l ) 1 6 / 5 4 A l b e r e t a l . i N N v e s t i g a t e N e u r a l N e t w o r k s , J M L R S o f t w a r e , 2 0 1 9

  17. A p p r o a c h 4 : G r a d i e n t x I n p u t I n p u t Mo d e l E x p l a n a t i o n V G G - 1 6 O b s e r v a t i o n : I n c e p t i o n V 3 E x p l a n a t i o n s a r e n o i s y . R e s N e t 5 0 1 7 / 5 4 A l b e r e t a l . i N N v e s t i g a t e N e u r a l N e t w o r k s , J M L R S o f t w a r e , 2 0 1 9

  18. A p p r o a c h 4 : G r a d i e n t x I n p u t T w o r e a s o n s w h y e x p l a n a t i o n s a r e n o i s y : N o t l o c a l e n o u g h . T o o m u c h c o n t e x t i n t r o d u c e d w h e n m u l t i p l y i n g b y t h e i n p u t . S h a t t e r e d g r a d i e n t p r o b l e m → g r a d i e n t o f d e e p n e t s h a s l o w i n f o r m a t i v e v a l u e 1 8 / 5 4

  19. A p p r o a c h 4 : G r a d i e n t x I n p u t The Shattered gradients problem [Montufar’14, Balduzzi’17] 1 9 / 5 4

  20. O v e r v i e w o f E x p l a n a t i o n M e t h o d s - R e c a p 1 . P e r t u r b a t i o n - B a s e d Me t h o d s → u n i v e r s a l l y a p p l i c a b l e b u t s l o w 2 . Me a n i n g f u l P e r t u r b a t i o n s → w i d e l y a p p l i c a b l e b u t r e q u i r e s o p t i m i z a t i o n 3 . T a y l o r E x p a n s i o n s → q u i t e w i d e l y a p p l i c a b l e b u t r e q u i r e s t o fj n d a r o o t p o i n t 4 . G r a d i e n t x I n p u t → a p p l i c a b l e w i t h s o m e r e s t r i c t i o n s → f a s t , O ( f o r w a r d p a s s ) → d o e s n o t w o r k w e l l o n h i g h l y n o n l i n e a r f u n c t i o n s ( e . g . D N N s ) 2 0 / 5 4

  21. L a y e r - Wi s e R e l e v a n c e P r o p a g a t i o n 2 1 / 5 4

  22. I d e a : R e u s i n g M o d e l S t r u c t u r e mo d e l i s a c o mp o s i t i o n o f n e u r o n s . T h i s c a n b e e x p l o i t e d t o ma k e e x p l a n a t i o n e a s i e r . 2 2 / 5 4

Recommend

EMBC Tutorial on Interpretable and Transparent Deep Learning Wojciech Samek Grgoire Montavon

EMBC Tutorial on Interpretable and Transparent Deep Learning Wojciech Samek Grgoire Montavon

EMBC Tutorial on Interpretable and Transparent Deep Learning Wojciech Samek Grgoire Montavon Klaus-Robert Mller (Fraunhofer HHI) (TU Berlin) (TU Berlin) 13:30 - 14:00 Introduction KRM 14:00 - 15:00 Techniques for Interpretability GM

1.03k views • 23 slides
From ML Successes to Applications ICIP18 Tutorial on Interpretable Deep Learning 2 Black Box

From ML Successes to Applications ICIP18 Tutorial on Interpretable Deep Learning 2 Black Box

From ML Successes to Applications ICIP18 Tutorial on Interpretable Deep Learning 2 Black Box Models Huge volumes of data Solve task Computing power Deep Neural Network Information (implicit) ICIP18 Tutorial on Interpretable Deep

540 views • 33 slides
Interpretable & Transparent Deep Learning Fraunhofer HHI, Machine Learning Group Wojciech

Interpretable & Transparent Deep Learning Fraunhofer HHI, Machine Learning Group Wojciech

Fraunhofer Image Processing Heinrich Hertz Institute Interpretable & Transparent Deep Learning Fraunhofer HHI, Machine Learning Group Wojciech Samek Northern Lights Deep Learning Workshop (NLDL19) Troms, Norway 8th January 2019

1.15k views • 51 slides
Data Fusion of Everything Tutorial at EMBC, Milan 2015 Tutorial Overview Motivation: search

Data Fusion of Everything Tutorial at EMBC, Milan 2015 Tutorial Overview Motivation: search

Marinka Zitnik & Blaz Zupan University of Ljubljana, Slovenia Data Fusion of Everything Tutorial at EMBC, Milan 2015 Tutorial Overview Motivation: search for bacterial-response genes, a case study Warm up: recommender systems

1.3k views • 68 slides
ICASSP 2017 Tutorial on Methods for Interpreting and Understanding Deep Neural Networks G.

ICASSP 2017 Tutorial on Methods for Interpreting and Understanding Deep Neural Networks G.

ICASSP 2017 Tutorial on Methods for Interpreting and Understanding Deep Neural Networks G. Montavon, W. Samek, K.-R. Mller Part 2: Making Deep Neural Networks Transparent 5 March 2017 Making Deep Neural Nets Transparent DNN transparency

448 views • 44 slides
Not Just a Black Box: Interpretable Deep Learning for Genomics Avan> Shrikumar, Peyton

Not Just a Black Box: Interpretable Deep Learning for Genomics Avan> Shrikumar, Peyton

Not Just a Black Box: Interpretable Deep Learning for Genomics Avan> Shrikumar, Peyton Greenside, Anshul Kundaje Peyton Anshul 1 With great power comes really poor interpretability Deep Interpretable Deep Power Learning Learning

783 views • 41 slides
Towards Interpretable Deep Learning for Natural Language Processing Roy Schwartz University of

Towards Interpretable Deep Learning for Natural Language Processing Roy Schwartz University of

Towards Interpretable Deep Learning for Natural Language Processing Roy Schwartz University of Washington & Allen Institute for Artificial Intelligence December 2018 1 / 37 ( Deep-Learning -Based) AI Today 2 / 37 3 / 37 Deep Learning

1.63k views • 140 slides
Toward In Interpretable De Deep Re Reinforcement Lea Learning g wi with Li Linea ear Model

Toward In Interpretable De Deep Re Reinforcement Lea Learning g wi with Li Linea ear Model

Toward In Interpretable De Deep Re Reinforcement Lea Learning g wi with Li Linea ear Model el U-Tr Trees Guiliang Liu, Oliver Schulte, Wang Zhu, Qingcan Li Machine Learning Lab, ECML-PKDD 2018 Presentation PROBLEM DEFINITION

372 views • 15 slides
Deep Visual Models with Interpretable Features and Modularized Structures Quanshi Zhang John

Deep Visual Models with Interpretable Features and Modularized Structures Quanshi Zhang John

Deep Visual Models with Interpretable Features and Modularized Structures Quanshi Zhang John Hopcroft Center Shanghai Jiao Tong University Quanshi Zhang, Ying Nian Wu, and Song-Chun Zhu, "Interpretable Convolutional Neural Networks"

694 views • 42 slides
outline of this tutorial motivations 1 ACISS09 tutorial on deep belief nets deep

outline of this tutorial motivations 1 ACISS09 tutorial on deep belief nets deep

outline of this tutorial motivations 1 ACISS09 tutorial on deep belief nets deep autoencoders deep belief nets sigmoid belief nets 2 Marcus Frean why are they hard to train? could layer-by-layer training work? Melbourne, 2009

509 views • 15 slides
Not Just a Black Box: Interpretable Deep Learning for Genomics Presented by: AvanA Shrikumar 1

Not Just a Black Box: Interpretable Deep Learning for Genomics Presented by: AvanA Shrikumar 1

Not Just a Black Box: Interpretable Deep Learning for Genomics Presented by: AvanA Shrikumar 1 With great power comes really poor interpretability Deep Power Learning Traditional machine learning Classical statistics 2

984 views • 82 slides
The INAC EMBC 1 Emergency Management Services Agreement - 2017 The INAC EMBC 2 Emergency

The INAC EMBC 1 Emergency Management Services Agreement - 2017 The INAC EMBC 2 Emergency

The INAC EMBC 1 Emergency Management Services Agreement - 2017 The INAC EMBC 2 Emergency Management Services Agreement - 2017 EMBC and INAC recently entered into a 10 year service agreement with INAC Effective date: April 1, 2017

480 views • 26 slides
SDRL: Interpretable and Data-efficient Deep Liu Reinforcement Learning Introduction Background

SDRL: Interpretable and Data-efficient Deep Liu Reinforcement Learning Introduction Background

SDRL: Symbolic Deep Reinforcement Learning SDRL: Interpretable and Data-efficient Deep Liu Reinforcement Learning Introduction Background Leveraging Symbolic Planning Method Experiment Conclusion Bo Liu and Future Work Auburn

1k views • 20 slides
Interpretable Multimodal Deep Learning for Objective Diagnosis, Prognosis and Biomarker Discovery

Interpretable Multimodal Deep Learning for Objective Diagnosis, Prognosis and Biomarker Discovery

Interpretable Multimodal Deep Learning for Objective Diagnosis, Prognosis and Biomarker Discovery Faisal Mahmood, PhD Postdoctoral Fellow Department of Biomedical Engineering Johns Hopkins University faisalm@jhu.edu | http://faisal.ai March

1.08k views • 62 slides
UCL Tutorial on: Deep Belief Nets (An updated and extended version of my 2007 NIPS tutorial)

UCL Tutorial on: Deep Belief Nets (An updated and extended version of my 2007 NIPS tutorial)

UCL Tutorial on: Deep Belief Nets (An updated and extended version of my 2007 NIPS tutorial) Geoffrey Hinton Canadian Institute for Advanced Research & Department of Computer Science University of Toronto Schedule for the Tutorial

1.66k views • 126 slides
Deep Learning Tutorial Part I Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning

Deep Learning Tutorial Part I Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning

Deep Learning Tutorial Part I Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning Tutorial,Part I 1 Overview Goals of the tutorial Somewhat organized overview of basics, and some more advanced topics Demistify jargon Pointers for

1.25k views • 105 slides
Deep Learning Tutorial Part II Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning

Deep Learning Tutorial Part II Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning

Deep Learning Tutorial Part II Greg Shakhnarovich TTI-Chicago December 2016 Deep Learning Tutorial,Part II 1 Overview Goals of the tutorial Somewhat organized overview of basics, and some more advanced topics Demistify jargon Pointers

925 views • 63 slides
with Interpretable Deep Learning Presented by: Avanti Shrikumar Kundaje Lab Stanford University

with Interpretable Deep Learning Presented by: Avanti Shrikumar Kundaje Lab Stanford University

Understanding Genome Regulation with Interpretable Deep Learning Presented by: Avanti Shrikumar Kundaje Lab Stanford University Example biological problem: understanding stem cell differentiation liver cells Lung cells fertilized egg

1.24k views • 74 slides
2019.07.04 Xie, P., Zuo, K., Zhang, Y., Li, F., Yin, M., & Lu, K. (2019). Interpretable

2019.07.04 Xie, P., Zuo, K., Zhang, Y., Li, F., Yin, M., & Lu, K. (2019). Interpretable

2019.07.04 Xie, P., Zuo, K., Zhang, Y., Li, F., Yin, M., & Lu, K. (2019). Interpretable Classification from Skin Cancer Histology Slides Ryan Chen Using Deep Learning: A Retrospective Multicenter Study. arXiv preprint arXiv:1904.06156.

488 views • 15 slides
Layer-wise Relevance Propagation in Neural Neural Networks Deep Learning Shortcomings Networks

Layer-wise Relevance Propagation in Neural Neural Networks Deep Learning Shortcomings Networks

LRP Ariyan Zarei Motivation Having More interpretable Layer-wise Relevance Propagation in Neural Neural Networks Deep Learning Shortcomings Networks to have more interpretable Papers and Demo Introduction Machine Learning models

373 views • 35 slides
Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 2

Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 2

Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 2 : M e t h o d s o f E x p l a n a t i o n W o j c i e c h S a m e k , G r g o i r e M o n t a v

475 views • 36 slides
Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 3

Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 3

Tutorials Interpretable Deep Learning: Towards Understanding & Explaining DNNs P a r t 3 : V a l i d a t i n g E x p l a n a t i o n s W o j c i e c h S a m e k , G r g o i r e M o n t

805 views • 28 slides
Deep Learning for Natural Language Inference NAACL-HLT 2019 Tutorial Follow the slides:

Deep Learning for Natural Language Inference NAACL-HLT 2019 Tutorial Follow the slides:

Deep Learning for Natural Language Inference NAACL-HLT 2019 Tutorial Follow the slides: nlitutorial.github.io Sam Bowman Xiaodan Zhu NYU (New York) Queens University, Canada Introduction Motivations of the Tutorial Overview Starting

2.13k views • 182 slides
4 Deep Generative Models BVM 2018 Tutorial: Advanced Deep Learning Methods Jens Petersen Dept.

4 Deep Generative Models BVM 2018 Tutorial: Advanced Deep Learning Methods Jens Petersen Dept.

4 Deep Generative Models BVM 2018 Tutorial: Advanced Deep Learning Methods Jens Petersen Dept. of Neuroradiology, Heidelberg University Hospital Div. of Medical Image Computing, DKFZ Heidelberg Faculty of Physics & Astronomy, Heidelberg

1.48k views • 65 slides

More recommend