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

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

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
Plug-and-play Integration of dual-model based Duftschmid Knowledge Artefacts into an Open Source

Plug-and-play Integration of dual-model based Duftschmid Knowledge Artefacts into an Open Source

Krexner, Plug-and-play Integration of dual-model based Duftschmid Knowledge Artefacts into an Open Source EHR Introduction openEHR Goal System Methods OpenMRS Form generation Export of EHR extracts Rabea Krexner Georg Duftschmid

770 views • 41 slides
R ETHINKING P YTHON P ACKAGING A thought experiment @pradyunsg pradyunsg.me/talks W HO ?

R ETHINKING P YTHON P ACKAGING A thought experiment @pradyunsg pradyunsg.me/talks W HO ?

R ETHINKING P YTHON P ACKAGING A thought experiment @pradyunsg pradyunsg.me/talks W HO ? Pradyun Gedam @pradyunsg pradyunsg.me Member of Python Packaging Authority Maintainer of pip, virtualenv, packaging and more PSF

804 views • 34 slides
Searches for Robotic Probes in the Solar System Eamonn Ansbro Kingsland Observatory RAS NAM:

Searches for Robotic Probes in the Solar System Eamonn Ansbro Kingsland Observatory RAS NAM:

Searches for Robotic Probes in the Solar System Eamonn Ansbro Kingsland Observatory RAS NAM: SETI Sessions St Andrews, July 5 2013 SETI, OSETI, SETA SETI, OSETI, SETA SETA: Search for Extraterrestrial Artefacts SETA: Search for

399 views • 35 slides
, Daniela Damian, Anthony Wasserman, Miguel Goulo, Joo Arajo Universidade Nova de

, Daniela Damian, Anthony Wasserman, Miguel Goulo, Joo Arajo Universidade Nova de

, Daniela Damian, Anthony Wasserman, Miguel Goulo, Joo Arajo Universidade Nova de Lisboa, Portugal University of Victoria, Canada Carnegie Mellon University, Silicon Valley, USA June 1st, 2018 2 Bob Dorf Steve Blank Eric

682 views • 30 slides
Software Development Methodologies Lecturer: Raman Ramsin Lecture 6 Integrated Object-Oriented

Software Development Methodologies Lecturer: Raman Ramsin Lecture 6 Integrated Object-Oriented

Software Development Methodologies Lecturer: Raman Ramsin Lecture 6 Integrated Object-Oriented Methodologies: RUP/USDP and EUP Sharif University of Technology Department of Computer Engineering 1 Software Development Methodologies

1.68k views • 26 slides
UNDERSTANDING LEADERSHIP STYLES Leadership on Demand AGENDA What is Leadership? What are

UNDERSTANDING LEADERSHIP STYLES Leadership on Demand AGENDA What is Leadership? What are

UNDERSTANDING LEADERSHIP STYLES Leadership on Demand AGENDA What is Leadership? What are Leadership Styles? Why think about leadership styles? Logic Problem Using True Colors TAKE OUT A PIECE OF PAPER 4:

375 views • 18 slides
Cambodia Leadership and Service Learning Journey Parent and Student Introduction Evening Service

Cambodia Leadership and Service Learning Journey Parent and Student Introduction Evening Service

Year 10 2019 Cambodia Leadership and Service Learning Journey Parent and Student Introduction Evening Service Learning and Leadership Journey Program Outcomes Courage To challenge your own view of yourself and others with the resilience

720 views • 16 slides
Automating Authority Work Automating authority work, or, Be your own authority control vendor

Automating Authority Work Automating authority work, or, Be your own authority control vendor

Mike Monaco Coordinator, Cataloging Services May 14, 2018 Automating Authority Work Automating authority work, or, Be your own authority control vendor Ohio Valley Group of Technical Services Librarians Mike Monaco 2018 Conference, May

1.06k views • 67 slides

More recommend