Towards multi-modal causability with Graph Neural Networks enabling information fusion for explainable AI
By Andreas Holzinger, Bernd Malle, Anna Saranti, Bastian Pfeifer. Information Fusion, 2021
307 citations recorded by the source corpus.
fields of multi-modal embeddings and interactive explainability, to lay the foundations for effective future human–AI interfaces. We emphasize that Graph Neural Networks play a major role for multi-modal causability, since causal links between features can be defined directly using graph structures.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/acc1150ac0b1b26257c1287084ae8b6ce380b8e3
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
- Przemyslaw Biecek
- Anna Saranti
- Andreas Holzinger
- Christoph Molnar
- Jun Xia
- Guang Yang
- Qinghao Ye
- Wojciech Samek
Citations in the corpus
- Graph Neural Networks for Intelligent Transportation Systems: A Survey (2023, background): “neural networks [215] can improve the robustness of the predictions.”
- Explainable AI Methods - A Brief Overview (2020, background): “Successful xAI models in the future will also require new human-AI interfaces [36] that enable contextual understanding and allow a domain expert to ask questions and counterfactuals [35] (”what-if” questions).”
- A Survey of Human-in-the-loop for Machine Learning (2021, methodology): “[140] adopted Graph Neural Networks as a method of choice, enabling information fusion for multimodal capability, which is a milestone.”
- Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond (2021, background): “It is of note that one most recent study advocated for the use of Graph Neural Networks as a tool of choice for multi-modal causability knowledge fusion [65].”