UI/UX Safety and Controls

  1. Checklist / Questions to ask your team

    • Do you have a mitigation plan for ML harm situations (ie. ystem that was designed and deployed for that task produced harmful and unexpected results) especially common with use of deep reinforcement and unsupervised learning, agents acting in broader environment settings.

    • Did you design to enable humans can step in and remedy if needed

    • How did you design not to make moves with very bad repercussions?

    • What happens if model has no training data for a specific scenario

    • Do you ensure privacy best practices when applying machine learning to sensitive data eg. medical data?

    2, Tools

    to shield from adversarial attacks the IBM Adversarial Robustness 360
    toolbox

    3. Use Cases

    4. Related Information

  2. Case Studies

    4. Further Readings