llms-as-orchestrators-constraint-compliant - Dual-agent optimization for constrained recommendations
Build DualAgent-Rec recommenders that optimize relevance, diversity, and novelty while enforcing hard constraints with dual-agent evolution and adaptive epsilon-relaxation.
Tags
Updated: 2026-09-30Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Define recommendation objectives
- Encode candidate lists
- Evaluate hard constraints
- Evolve dual populations
- Transfer elite solutions
- Coordinate agent allocation
- Relax constraints adaptively
- Select feasible recommendations
Inputs
- User history
- Item catalog
- Item metadata
- Base relevance scores
- Objective definitions
- Constraint thresholds
- Recommendation list length
- User preferences
Outputs
- Feasible recommendation lists
- Pareto-optimal recommendation sets
- Constraint satisfaction metrics
- Optimization state metrics
Requirements
- Python execution environment
- LLM coordinator access
- Pre-trained item embeddings