Toward Trustworthy Learning-Enabled Systems with Concept-Based Explanations
Despite the superior performance of deep learning-based controllers in network applications, their practical adoption is limited due to the difficulty in understanding and trusting them. Existing explainability solutions largely focus on interpreting these controllers by providing insights into the top features used by the model. Although these insights can help reveal an important aspect of the controller, they require operators to deal with low-level features, requiring extensive manual analysis and interpretation.
Authors
- Dongsu Han (ORCID: https://orcid.org/0000-0001-6922-7244)
- Nina Narodytska (ORCID: https://orcid.org/0000-0002-5181-4560)
- Sangeetha Abdu Jyothi (ORCID: https://orcid.org/0009-0000-0503-4478)
- Sagar Patel
Institutions
- Broadcom (Israel) (IL)
- Korea Advanced Institute of Science and Technology (KR)
- University of California, Irvine (US)
- Kootenay Association for Science & Technology (CA)
- Broadcom (United States) (US)
Publication Details
- Published
- 2024-11-11
- DOI
- https://doi.org/10.1145/3696348.3696894
- Citations
- 3
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 1.04
Funders
- Ministry of Science and ICT, South Korea
- Institute for Information and Communications Technology Promotion