methods for interpreting and understanding deep neural
play

Methods for interpreting and understanding deep neural networks - PowerPoint PPT Presentation

Methods for interpreting and understanding deep neural networks Grgoire Montavon, Wojciech Samek, Klaus-Robert Mller 2017 Presented by Philipp Wimmer M o t i v a t i o n U n d e r s t a n d i n g a n d v


  1. „Methods for interpreting and understanding deep neural networks“ Grégoire Montavon, Wojciech Samek, Klaus-Robert Müller 2017 Presented by Philipp Wimmer

  2. M o t i v a t i o n ● U n d e r s t a n d i n g a n d v a l i d a t i n g d e e p n e u r a l n e t w o r k s i s h a r d – M a n y p a r a m e t e r s – H i g h l y n o n l i n e a r – I n t e r p r a t a b i l i t y w a s n ‘ t a g o a l o f D N N s ● A b i l i t y t o v a l i d a t e i s n e c c e s s a r y f o r u n d e r s t a n d i n g a n d r e a l w o r l d a p p l i c a b i l i t y ● E x a m p l e : D o n ‘ t k n o w i f h i g h p r e d i c t i o n a c c u r a c y i s d u e t o a n o m a l y i n t r a i n i n g d a t a

  3. I n t e r p r e t a t i o n A n i n t e r p r e t a t i o n i s t h e m a p p i n g o f a n a b s t r a c t c o n c e p t ( e . g . a p r e d i c t e d c l a s s ) i n t o a d o m a i n t h a t t h e h u m a n c a n m a k e s e n s e o f . G o a l : P r o d u c i n g a p r o t o t y p e

  4. E x p l a n a t i o n A n e x p l a n a t i o n i s t h e c o l l e c t i o n o f f e a t u r e s o f t h e i n t e r p r e t a b l e d o m a i n , t h a t h a v e c o n t r i b u t e d f o r a g i v e n e x a m p l e t o p r o d u c e a d e c i s i o n ( e . g . c l a s s i fi c a t i o n o r r e g r e s s i o n ) . G o a l : P r o d u c i n g a h e a t ma p

  5. P a r t A : I n t e r p r e t i n g ● A c t i v a t i o n M a x i m a z a t i o n ( A M ) ● A M w i t h a n e x p e r t ● A M i n c o d e s p a c e ( u s i n g G e n e r a t i v e A d v e r s e r i a l N e t w o r k s )

  6. A c t i v a t i o n M a x i m i z a t i o n ● P r o d u c i n g a p r o t o t y p e v i a m a x i m i z i n g C l a s s p r o b a b i l i t i e s R e g u l a r i z e r ● C l a s s p r o b a b i l i t e s m o d e l e d b y t h e D N N a r e f u n c t i o n s w i t h a g r a d i e n t ● U s e g r a d i e n t d e s c e n t t o m a x i m i z e ( j u s t l i k e t r a i n i n g a D N N i n r e v e r s e

  7. A r c h i t e c t u r e o f A M ● S i m p l e t o c o m p u t e ● R e g u l a r i z e r p r e f e r e s i n p u t s c l o s e t o t h e o r i g i n ( m e a n o f d a t a ) ● U n n a t u r a l l o o k i n g p r o t o y p e

  8. I m p r o v i n g A M w i t h e x p e r t ● R e p l a c e r e g u l a r i z e r w i t h a m o r e s o p h i s t i c a t e d a p p r o a c h C l a s s p r o b a b i l i t i e s M o d e l o f t h e d a t a ● E x p e r t i s t h e d a t a d e n s i t y ● F o r e x a m p l e o b t a i n e d b y t r a i n i n g a n G a u s s i a n R B M ● O f t e n m o r e c o m p l e x d e n s i t y m o d e l s a r e n e e d e d

  9. (a) maximation of class probability function (b) favoring natural images – often suffjcient. (c) desired (d) optimization of the expert itself, hides failure modes

  10. P e r f o r m i n g A M i n c o d e S p a c e ● O f t e n l e a r n i n g t h e e x p e r t t o a h i g h a c c u r a c y i s h a r d ● E x p e r t o f t e n v e r y c o m p l e x s u c h t h a t m a x i m i z i n g i s d i ffi c u l t ● A s o l u t i o n i s t o n o t e x p l i c i t e l y l e a r n p ( x ) ● I n s t e a d s a m p l e f r o m a n c o d e s p a c e w i t h k n o w n d i s t r i b u t i o n w h i c h w a s o b t a i n e d b y t r a i n i n g a n G A N D e c o d e d p o i n t i n C o d e s p a c e ● T h e n a p p l y d e c o d i n g f u n c t i o n t o g e t a p r o t o y p e

  11. P e r f o r m i n g A M i n c o d e s p a c e ● D i s t r i b u t i o n i n c o d e s p a c e i s b y c o n s t r u c t i o n G a u s s i a n ● R e g u l a r i z e r f a v o r s p o i n t s w i t h a h i g h p r o b a b i l i t y

  12. R e s u l t s S i m p l e A M A M w i t h e x p e r t A M i n c o d e s p a c e

  13. P a r t B : E x p l a i n i n g ● S e n s i t i v i t y A n a l y s i s ● ( L a y e r w i s e ) R e l e v a n c e P r o p a g a t i o n

  14. S e n s i t i v i t y A n a l y s i s

  15. S e n s i t i v i t y A n a l y s i s ● G r a d i e n t i s e a s i l y c a l c u l a t e d v i a b a c k p r o p a g a t i o n ● T h e m e a s u r e d r e l e v a n c e s c o r e i s n o t w h a t i s w a n t e d ● M e a s u r e s n o t t h e r e l e v a n c e , b u t t h e l o c a l s l o p e o f i t

  16. R e l e v a n c e P r o p a g a t i o n ● M a k e u s e o f t h e g r a p h s t r u c t u r e o f D N N s ● P r o p a g a t e t h e r e l e v a n c e s c o r e b a c k w a r d s t h r o u g h t h e n e t w o r k i s s i m i l a r t o t h e b a c k p r o p a g a t i o n o f t h e e r r o r d u r i n g t h e t r a i n i n g p h a s e ● R e l e v a n c e h a s t o b e c o n s e r v e d ( s i m i l a r t o c u r r e n t i n a n e l e c t r i c c i r c u i t ) ● L o c a l c o n s e r v a t i o n a t e a c h n e u r o n ● F i l t e r i n g : A b l e t o b l o c k t h e fl o w t h r o u g h c e r t a i n n e u r o n s

  17. P r o p a g a t i o n R u l e C o u n t e r - R e l e v a n c e R e l e v a n c e

  18. H y p e r p a r a m e t e r s o f L R P ● R a t i o o f α a n d β d e t e r m i n e s t h e i n fl u e n c e o f c o u n t e r v a r i a n c e

  19. C o n c l u s i o n ● T w o m e t h o d s f o r i n c r e a s i n g p o s t - h o c i n t e r p r e t a b i l i t y ● N o n e e d t o c h a n g e e x i s t i n g a l g o r i t h m s ● E n a b l e s b e t t e r u n d e r s t a n d i n g a n d v a l i d a t i o n ● S h o u l d b e i n t h e t o o l b o x o f e v e r y o n e u s i n g D N N s

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


More recommend