Explainable AI: Interpreting, Explaining and Visualizing Deep Learning

By Wojciech Samek, Grgoire Montavon, Andrea Vedaldi, Lars Kai Hansen, Klaus‐Robert Müller . Explainable AI, 2019

1,312 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/0e1f153576c7f9f2628cdd34a1067c4d26bdc096

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Multi-Modal Knowledge Graph Construction and Application: A Survey (2022, methodology): “Saliency-based methods treat the marginal effects [63] of pixels to a given phrase by gradient computation as the heatmap value, and attention-based methods treat the cross-modal relevance as the heatmap value.”
Interpretable Machine Learning (2019, background): “90,91 It works by creating a visual map highlighting the parts of the image (or other data) that most influenced the ML system’s output.”
TransPose: Keypoint Localization via Transformer (2020, background): “As surveyed by [47], many works define the goal for explanation is to determine what inputs are the most relevant to the prediction, which is also the goal we seek in this paper.”
Evaluating the Quality of Machine Learning Explanations: A Survey on Methods and Metrics (2021, background): “The challenge is that there is no ground truth when evaluating post-hoc explanations, as the real inner workings of the model is not known [60].”
Physics-informed machine learning: case studies for weather and climate modelling (2021, background): “While there exists a large body of literature in interpretable ML [75–77], Rudin [78] argues that because interpretability needs to be deﬁned in a domain-speciﬁc way, some of the most important technical challenges for the future of interpretable ML models are tied to the needs of speciﬁc domains.”
Artificial Intelligence For Natural Product Drug Discovery (2023, background): “…contain crucial information about each structure (for example, SMILES, compound name, availability and location of spectral data, source organism and BGC) and would offer a central point of reference for data dissemination and automated database importation by natural product-centric resources.”
Benchmarking and survey of explanation methods for black box models (2021, background): “Another branch of the literature review focuses on the quantitative and qualitative evaluation of explanation methods (Samek et al. 2019; Carvalho et al. 2019).”
You Shouldn't Trust Me: Learning Models Which Conceal Unfairness From Multiple Explanation Methods (2020, background): “One motivation is to ensure fairness as part of the ‘Fair, Accountable, and Transparent’ research agenda [9, 36].”
What Role Does Hydrological Science Play in the Age of Machine Learning? (2020, background): “First, we can leverage advances in explainable AI (XAI; Samek, 2019).”