XAI-TRIS: non-linear image benchmarks to quantify false positive post-hoc attribution of feature importance

Benedict Clark, Rick Wilming, Stefan Haufe. XAI-TRIS: non-linear image benchmarks to quantify false positive post-hoc attribution of feature importance. Machine Learning, 113(9):6871-6910, September 2024. [doi]

Authors

Benedict Clark

This author has not been identified. Look up 'Benedict Clark' in Google

Rick Wilming

This author has not been identified. Look up 'Rick Wilming' in Google

Stefan Haufe

This author has not been identified. Look up 'Stefan Haufe' in Google