Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.14338 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915939424927744 |
|---|---|
| author | Kamalov, Firuz Thabtah, Fadi Sivaraj, R. Abdelhamid, Neda |
| author_facet | Kamalov, Firuz Thabtah, Fadi Sivaraj, R. Abdelhamid, Neda |
| contents | We introduce path-sampled integrated gradients (PS-IG), a framework that generalizes feature attribution by computing the expected value over baselines sampled along the linear interpolation path. We prove that PS-IG is mathematically equivalent to path-weighted integrated gradients, provided the weighting function matches the cumulative distribution function of the sampling density. This equivalence allows the stochastic expectation to be evaluated via a deterministic Riemann sum, improving the error convergence rate from $O(m^{-1/2})$ to $O(m^{-1})$ for smooth models. Furthermore, we demonstrate analytically that PS-IG functions as a variance-reducing filter against gradient noise - strictly lowering attribution variance by a factor of 1/3 under uniform sampling - while preserving key axiomatic properties such as linearity and implementation invariance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_14338 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Path-Sampled Integrated Gradients Kamalov, Firuz Thabtah, Fadi Sivaraj, R. Abdelhamid, Neda Machine Learning We introduce path-sampled integrated gradients (PS-IG), a framework that generalizes feature attribution by computing the expected value over baselines sampled along the linear interpolation path. We prove that PS-IG is mathematically equivalent to path-weighted integrated gradients, provided the weighting function matches the cumulative distribution function of the sampling density. This equivalence allows the stochastic expectation to be evaluated via a deterministic Riemann sum, improving the error convergence rate from $O(m^{-1/2})$ to $O(m^{-1})$ for smooth models. Furthermore, we demonstrate analytically that PS-IG functions as a variance-reducing filter against gradient noise - strictly lowering attribution variance by a factor of 1/3 under uniform sampling - while preserving key axiomatic properties such as linearity and implementation invariance. |
| title | Path-Sampled Integrated Gradients |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.14338 |