Making deep neural networks right for the right scientific reasons by interacting with their explanations
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2020
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916146817531904 |
|---|---|
| author | Schramowski, Patrick Stammer, Wolfgang Teso, Stefano Brugger, Anna Shao, Xiaoting Luigs, Hans-Georg Mahlein, Anne-Katrin Kersting, Kristian |
| author_facet | Schramowski, Patrick Stammer, Wolfgang Teso, Stefano Brugger, Anna Shao, Xiaoting Luigs, Hans-Georg Mahlein, Anne-Katrin Kersting, Kristian |
| contents | Deep neural networks have shown excellent performances in many real-world applications. Unfortunately, they may show "Clever Hans"-like behavior -- making use of confounding factors within datasets -- to achieve high performance. In this work, we introduce the novel learning setting of "explanatory interactive learning" (XIL) and illustrate its benefits on a plant phenotyping research task. XIL adds the scientist into the training loop such that she interactively revises the original model via providing feedback on its explanations. Our experimental results demonstrate that XIL can help avoiding Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust into the underlying model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2001_05371 |
| institution | arXiv |
| publishDate | 2020 |
| record_format | arxiv |
| spellingShingle | Making deep neural networks right for the right scientific reasons by interacting with their explanations Schramowski, Patrick Stammer, Wolfgang Teso, Stefano Brugger, Anna Shao, Xiaoting Luigs, Hans-Georg Mahlein, Anne-Katrin Kersting, Kristian Machine Learning Artificial Intelligence Deep neural networks have shown excellent performances in many real-world applications. Unfortunately, they may show "Clever Hans"-like behavior -- making use of confounding factors within datasets -- to achieve high performance. In this work, we introduce the novel learning setting of "explanatory interactive learning" (XIL) and illustrate its benefits on a plant phenotyping research task. XIL adds the scientist into the training loop such that she interactively revises the original model via providing feedback on its explanations. Our experimental results demonstrate that XIL can help avoiding Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust into the underlying model. |
| title | Making deep neural networks right for the right scientific reasons by interacting with their explanations |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2001.05371 |