Global Explanations for Image Classifiers (Student Abstract)

AAAI 2023 pp. 16352-16353

doi:10.1609/AAAI.V37I13.27036 /aaai/2023/vasu2023aaai-global/

Abstract

We hypothesize that deep network classifications of complex scenes can be explained using sets of relevant objects. We employ beam search and singular value decomposition to generate local and global explanations that summarize the deep model's interpretation of a class.

PDF AAAI Semantic Scholar

Cite

Text

Vasu and Tadepalli. "Global Explanations for Image Classifiers (Student Abstract)." AAAI Conference on Artificial Intelligence, 2023. doi:10.1609/AAAI.V37I13.27036

Markdown

[Vasu and Tadepalli. "Global Explanations for Image Classifiers (Student Abstract)." AAAI Conference on Artificial Intelligence, 2023.](https://mlanthology.org/aaai/2023/vasu2023aaai-global/) doi:10.1609/AAAI.V37I13.27036

BibTeX

@inproceedings{vasu2023aaai-global,
  title     = {{Global Explanations for Image Classifiers (Student Abstract)}},
  author    = {Vasu, Bhavan K. and Tadepalli, Prasad},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2023},
  pages     = {16352-16353},
  doi       = {10.1609/AAAI.V37I13.27036},
  url       = {https://mlanthology.org/aaai/2023/vasu2023aaai-global/}
}