Abstract | ||
---|---|---|
Explaining the output of a deep network remains a challenge. In the case of an image classifier, one type of explanation is to identify pixels that strongly influence the final decision. A starting point for this strategy is the gradient of the class score function with respect to the input image. This gradient can be interpreted as a sensitivity map, and there are several techniques that elaborate on this basic idea. This paper makes two contributions: it introduces SmoothGrad, a simple method that can help visually sharpen gradient-based sensitivity maps, and it discusses lessons in the visualization of these maps. We publish the code for our experiments and a website with our results. |
Year | Venue | Field |
---|---|---|
2017 | arXiv: Learning | Publication,Image classifier,Data mining,Computer science,Visualization,Artificial intelligence,Pixel,Score,Machine learning |
DocType | Volume | Citations |
Journal | abs/1706.03825 | 51 |
PageRank | References | Authors |
1.41 | 2 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daniel Smilkov | 1 | 87 | 5.81 |
Nikhil Thorat | 2 | 92 | 3.54 |
Been Kim | 3 | 353 | 21.44 |
Fernanda B. Viégas | 4 | 3208 | 283.62 |
Martin Wattenberg | 5 | 4695 | 333.69 |