Explainable AI in Drug Discovery

  • Explainable AI in Drug Discovery

    Posted by Arthur on January 17, 2025 at 12:09 pm

    We’re working on improving the explainability of our AI models in drug discovery. Any recommendations on balancing model performance with interpretability?

    Grace replied 1 month, 3 weeks ago 3 Members · 2 Replies
  • 2 Replies
  • Sophia

    Member
    January 17, 2025 at 12:12 pm

    That’s a critical area, Arthur. We’ve been exploring SHAP (SHapley Additive exPlanations) values to interpret our models. It’s been particularly useful in understanding feature importance in molecular property prediction.

  • Grace

    Member
    January 17, 2025 at 12:14 pm

    This is an important area from a regulatory standpoint. We’re considering guidelines that would require a certain level of explainability for AI models used in drug discovery. What metrics do you think would be appropriate for measuring explainability?

Log in to reply.

});