Feedback Grading

Overview

Feedback Grading allows human operators to review and grade OmegaEngine decisions. This creates a continuous learning loop that improves policy accuracy over time.

• ✓ Grade decisions as correct, incorrect, or needs review
• ✓ Add notes and context to each grade
• ✓ Automatic policy weight adjustment based on feedback
• ✓ Track grader accuracy over time

API Example

// Submit feedback for a decision
await omega.feedback.grade({
  decisionId: "dec_abc123",
  grade: "INCORRECT",
  correctOutcome: "ALLOW",
  reason: "False positive - vendor was already approved",
  graderId: "operator@company.com"
});

// Bulk feedback for training
await omega.feedback.bulkGrade([
  { decisionId: "dec_1", grade: "CORRECT" },
  { decisionId: "dec_2", grade: "CORRECT" },
  { decisionId: "dec_3", grade: "INCORRECT", correctOutcome: "BLOCK" }
]);