Interface in Azure Machine Learning that brings together error analysis, fairness, interpretability, data analysis and counterfactual and causal analysis, helping you debug and assess a model ahead of deployment.
Also called RAI dashboard.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Responsible AI dashboard in context, with comparison tables and the common traps.
Terms in this definition
- Azure Machine Learning
Azure platform where you train, deploy, monitor and retrain models of your own, practising MLOps through tools such as pipelines, online endpoints, model monitoring and prompt flow.
- Error analysis
A Responsible AI dashboard tool. Using a decision tree and a heat map, it pinpoints cohorts that err more often than the model does overall.
- Model deployment
Inside a resource, each deployment is a named copy of a Foundry model with its own TPM quota and deployment type. Requests identify the model by this deployment name.
Related terms
- Cohort
In the Responsible AI dashboard, a group of data points you pick out so you can set its errors and metrics against others and spot gaps that one global cohort conceals.
- Fairness assessment
Responsible AI dashboard tool, based on Fairlearn, that measures gaps in model performance and selection rate between groups formed from sensitive features like age, ethnicity or gender.