🧪 Review Suggested Experiments
Focus question: What experiments should I run next to optimize based on my goals?
GOAL: [[[TBD: Add dynamic text that will change based on the optimization goal. Including it right above the table will remind users of what they set.]]]
Suggested Experiments
(Experiments expected to give the greatest improvement towards your optimization goal)
Note: TBD Use values from acquisition function to color code experiments as diverse (exploratory) or exploitative (near the current best).
Recalculate Suggested Experiments
(Re-rank Suggestions Based on Updated Weights)
TBD Visualize Suggested Experiments
(Partial Dependence Acquisition Plots)
TBD Visualize Multi-Objective Trade-offs
(Categorical Grid with Multi-Objective BI Plot)
Time Profiles of Suggestions
(Line Graph of Predicted Outputs Over Time)
TBD Best-So-Far Tracker
(Performance of Top Suggestions Across Iterations)
Long-term goals: Allow the optimization module to be rerun independently of data and ML workflows. This would let users change between optimization goals, use advanced weight-based and reference-point-based optimization methods, and avoid rerunning the entire data and modeling workflows.