MM-INTEREST Archives

ACM SIGMM Interest List

MM-INTEREST@LISTSERV.ACM.ORG

Options: Use Classic View

Use Monospaced Font
Show HTML Part by Default
Condense Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.14\))
Sender: ACM SIGMM Interest List <[log in to unmask]>
From: jbenoisp <[log in to unmask]>
Date: Tue, 28 Feb 2023 17:27:42 +0100
Message-ID: <[log in to unmask]>
Content-Type: multipart/alternative; boundary="Apple-Mail=_E25D582E-B2D8-481B-862B-B7D4944FC499"
Reply-To: jbenoisp <[log in to unmask]>
Parts/Attachments: text/plain (2383 bytes) , text/html (4 kB)
Dear all, 
Our method FEM (Feature Explanation Method) for sample-based explanations of DNN décisions in Image Classification tasks has been released in OpenSource 
You can find it here 
https://github.com/labribkb/fem <https://github.com/labribkb/fem> 

The summary description of it available in Open source is in here : Alexey  Zhukov, Jenny Benois-Pineau, Romain Giot, «Evaluation of Explanation Methods of AI - CNNs in Image
Classification Tasks with Reference-based and No-reference Metrics », in Advances in Artificial Intelligence and Machine Learning; Research 2 (4) 620-646, Article in Publishing  
https://www.oajaiml.com/uploads/archivepdf/20401143.pdf <https://www.oajaiml.com/uploads/archivepdf/20401143.pdf>

Its application to medical images is in the chapter n6  of the newly published Book Explainable Deep Learning - AI: Methods and Challenges, Elsevier , 2023
https://www.elsevier.com/books/explainable-deep-learning-ai/benois-pineau/978-0-323-96098-4 <https://www.elsevier.com/books/explainable-deep-learning-ai/benois-pineau/978-0-323-96098-4>; 

The methods works without gradient computation but by a statistical analysis. 


It is easy to play with, we thank you in advance for your feed-back. 

Jenny Benois-Pineau 
on behalf of the authors. 

Jenny Benois-Pineau, 
Professeure en Informatique, 
Chargée de mission aux relations Internationales
Collège Sciences et Technologies, 
Université de Bordeaux
351, crs de la Libération
33405 Talence
France
tel.: +33 (0) 5 40 00 84 24

Jenny Benois-Pineau, PhD, HDR, 
Professor of Computer Science, 
Chair of International relations
School of Sciences and Technologies
University of Bordeaux 
351, crs de la Libération
33405 Talence
tel.: +33 (0) 5 40 00 84 24


############################

Unsubscribe:

[log in to unmask]

If you don't already have a password for the LISTSERV.ACM.ORG server, we recommend
that you create one now. A LISTSERV password is linked to your email
address and can be used to access the web interface and all the lists to
which you are subscribed on the LISTSERV.ACM.ORG server.

To create a password, visit:

https://LISTSERV.ACM.ORG/SCRIPTS/WA-ACMLPX.CGI?GETPW1

Once you have created a password, you can log in and view or change your
subscription settings at:

https://LISTSERV.ACM.ORG/SCRIPTS/WA-ACMLPX.CGI?SUBED1=MM-INTEREST


ATOM RSS1 RSS2