Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929535283363840 |
|---|---|
| author | Bareeva, Dilyara Yolcu, Galip Ümit Hedström, Anna Schmolenski, Niklas Wiegand, Thomas Samek, Wojciech Lapuschkin, Sebastian |
| author_facet | Bareeva, Dilyara Yolcu, Galip Ümit Hedström, Anna Schmolenski, Niklas Wiegand, Thomas Samek, Wojciech Lapuschkin, Sebastian |
| contents | In recent years, training data attribution (TDA) methods have emerged as a promising direction for the interpretability of neural networks. While research around TDA is thriving, limited effort has been dedicated to the evaluation of attributions. Similar to the development of evaluation metrics for traditional feature attribution approaches, several standalone metrics have been proposed to evaluate the quality of TDA methods across various contexts. However, the lack of a unified framework that allows for systematic comparison limits trust in TDA methods and stunts their widespread adoption. To address this research gap, we introduce Quanda, a Python toolkit designed to facilitate the evaluation of TDA methods. Beyond offering a comprehensive set of evaluation metrics, Quanda provides a uniform interface for seamless integration with existing TDA implementations across different repositories, thus enabling systematic benchmarking. The toolkit is user-friendly, thoroughly tested, well-documented, and available as an open-source library on PyPi and under https://github.com/dilyabareeva/quanda. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_07158 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond Bareeva, Dilyara Yolcu, Galip Ümit Hedström, Anna Schmolenski, Niklas Wiegand, Thomas Samek, Wojciech Lapuschkin, Sebastian Machine Learning Artificial Intelligence In recent years, training data attribution (TDA) methods have emerged as a promising direction for the interpretability of neural networks. While research around TDA is thriving, limited effort has been dedicated to the evaluation of attributions. Similar to the development of evaluation metrics for traditional feature attribution approaches, several standalone metrics have been proposed to evaluate the quality of TDA methods across various contexts. However, the lack of a unified framework that allows for systematic comparison limits trust in TDA methods and stunts their widespread adoption. To address this research gap, we introduce Quanda, a Python toolkit designed to facilitate the evaluation of TDA methods. Beyond offering a comprehensive set of evaluation metrics, Quanda provides a uniform interface for seamless integration with existing TDA implementations across different repositories, thus enabling systematic benchmarking. The toolkit is user-friendly, thoroughly tested, well-documented, and available as an open-source library on PyPi and under https://github.com/dilyabareeva/quanda. |
| title | Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2410.07158 |