Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.11729 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909227571740672 |
|---|---|
| author | Machiraju, Gautam Derry, Alexander Desai, Arjun Guha, Neel Karimi, Amir-Hossein Zou, James Altman, Russ Ré, Christopher Mallick, Parag |
| author_facet | Machiraju, Gautam Derry, Alexander Desai, Arjun Guha, Neel Karimi, Amir-Hossein Zou, James Altman, Russ Ré, Christopher Mallick, Parag |
| contents | Feature attribution, the ability to localize regions of the input data that are relevant for classification, is an important capability for ML models in scientific and biomedical domains. Current methods for feature attribution, which rely on "explaining" the predictions of end-to-end classifiers, suffer from imprecise feature localization and are inadequate for use with small sample sizes and high-dimensional datasets due to computational challenges. We introduce prospector heads, an efficient and interpretable alternative to explanation-based attribution methods that can be applied to any encoder and any data modality. Prospector heads generalize across modalities through experiments on sequences (text), images (pathology), and graphs (protein structures), outperforming baseline attribution methods by up to 26.3 points in mean localization AUPRC. We also demonstrate how prospector heads enable improved interpretation and discovery of class-specific patterns in input data. Through their high performance, flexibility, and generalizability, prospectors provide a framework for improving trust and transparency for ML models in complex domains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_11729 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Prospector Heads: Generalized Feature Attribution for Large Models & Data Machiraju, Gautam Derry, Alexander Desai, Arjun Guha, Neel Karimi, Amir-Hossein Zou, James Altman, Russ Ré, Christopher Mallick, Parag Machine Learning Artificial Intelligence Quantitative Methods Feature attribution, the ability to localize regions of the input data that are relevant for classification, is an important capability for ML models in scientific and biomedical domains. Current methods for feature attribution, which rely on "explaining" the predictions of end-to-end classifiers, suffer from imprecise feature localization and are inadequate for use with small sample sizes and high-dimensional datasets due to computational challenges. We introduce prospector heads, an efficient and interpretable alternative to explanation-based attribution methods that can be applied to any encoder and any data modality. Prospector heads generalize across modalities through experiments on sequences (text), images (pathology), and graphs (protein structures), outperforming baseline attribution methods by up to 26.3 points in mean localization AUPRC. We also demonstrate how prospector heads enable improved interpretation and discovery of class-specific patterns in input data. Through their high performance, flexibility, and generalizability, prospectors provide a framework for improving trust and transparency for ML models in complex domains. |
| title | Prospector Heads: Generalized Feature Attribution for Large Models & Data |
| topic | Machine Learning Artificial Intelligence Quantitative Methods |
| url | https://arxiv.org/abs/2402.11729 |