Deep neural networks have demonstrated excellent performances in many real-world applications. Unfortunately, they may show Clever Hans-like behaviour (making use of confounding factors within datasets) to achieve high performance. In this work we introduce the novel learning setting of explanatory interactive learning and illustrate its benefits on a plant phenotyping research task. Explanatory interactive learning adds the scientist into the training loop, who interactively revises the original model by providing feedback on its explanations. Our experimental results demonstrate that explanatory interactive learning can help to avoid Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust in the underlying model.

Making deep neural networks right for the right scientific reasons by interacting with their explanations / Schramowski, Patrick; Stammer, Wolfgang; Teso, Stefano; Brugger, Anna; Herbert, Franziska; Shao, Xiaoting; Luigs, Hans-Georg; Mahlein, Anne-Katrin; Kersting, Kristian. - In: NATURE MACHINE INTELLIGENCE. - ISSN 2522-5839. - 2:8(2020), pp. 476-486. [10.1038/s42256-020-0212-3]

Making deep neural networks right for the right scientific reasons by interacting with their explanations

Teso, Stefano;
2020-01-01

Abstract

Deep neural networks have demonstrated excellent performances in many real-world applications. Unfortunately, they may show Clever Hans-like behaviour (making use of confounding factors within datasets) to achieve high performance. In this work we introduce the novel learning setting of explanatory interactive learning and illustrate its benefits on a plant phenotyping research task. Explanatory interactive learning adds the scientist into the training loop, who interactively revises the original model by providing feedback on its explanations. Our experimental results demonstrate that explanatory interactive learning can help to avoid Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust in the underlying model.
2020
8
Schramowski, Patrick; Stammer, Wolfgang; Teso, Stefano; Brugger, Anna; Herbert, Franziska; Shao, Xiaoting; Luigs, Hans-Georg; Mahlein, Anne-Katrin; Ke...espandi
Making deep neural networks right for the right scientific reasons by interacting with their explanations / Schramowski, Patrick; Stammer, Wolfgang; Teso, Stefano; Brugger, Anna; Herbert, Franziska; Shao, Xiaoting; Luigs, Hans-Georg; Mahlein, Anne-Katrin; Kersting, Kristian. - In: NATURE MACHINE INTELLIGENCE. - ISSN 2522-5839. - 2:8(2020), pp. 476-486. [10.1038/s42256-020-0212-3]
File in questo prodotto:
File Dimensione Formato  
nat mach intell.pdf

Open Access dal 13/02/2021

Descrizione: Articolo principale
Tipologia: Post-print referato (Refereed author’s manuscript)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 334.98 kB
Formato Adobe PDF
334.98 kB Adobe PDF Visualizza/Apri
nat mach intell supplementary.pdf

Open Access dal 13/02/2021

Descrizione: Materiale supplementare
Tipologia: Altro materiale allegato (Other attachments)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 3.4 MB
Formato Adobe PDF
3.4 MB Adobe PDF Visualizza/Apri
s42256-020-0212-3.pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 6.61 MB
Formato Adobe PDF
6.61 MB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/290515
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 121
  • ???jsp.display-item.citation.isi??? 84
social impact