AI engineering glossary · AI ethics law product
What is saliency map?
Saliency map: A visualisation, mainly for image models, that highlights which pixels most influence a prediction, usually by computing the gradient of the output with respect to the input. It helps check whether a model looks at the right region, but can be noisy.
Saliency map: an example
A saliency map of a chest X-ray classifier lights up a text marker in the corner instead of the lungs, revealing a shortcut the model learned.
Related terms
- Grad-CAM
- Integrated gradients
- Post-hoc explanation
- Explanation faithfulness
More AI ethics law product terms
- Sensitive personal data
- Shadow mode
- SHAP
- Significant Data Fiduciary
- Transparency (AI)
- Trust calibration
- Verifiable parental consent
- Watermarking AI outputs
Where you learn it
Saliency map is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.