yogsoth-ai Efficient Exploration
Purpose
Produce reliable rankings when the candidate set is too large for complete comparison. Uses information-theoretic pair selection and sparse-matrix rating algorithms to converge quickly with minimal comparisons.
When to use
- Candidate count N ≥ 100
- Complete comparison infeasible (budget << N(N-1)/2)
- Approximate ranking acceptable — top-k identification sufficient
- Speed/efficiency prioritized over perfect calibration
Budget
| Resource | Allocation |
|---|---|
| Comparisons | N×log(N) to 3N×log(N) |
| Iterations | 5-20 rounds of adaptive selection |
| Convergence target | Top-k stability ≥ 90% for 3 consecutive rounds |
State Ledger
candidates: [] # full candidate list
comparison_history: [] # [{pair, winner, confidence, round}]
ratings: {} # candidate → {mu, sigma}
method: "" # trueskill | bt-incomplete | rank-centrality
iteration: 0
budget_remaining: 0
convergence: {stable: false, score: 0.0, top_k_stable: false}
Available Tactics
- adaptive-pair-selection — maximize information gain per comparison
- consistency-audit-loop — spot-check transitivity in top-k region
Available SOPs
- pair-selector
- comparison-executor
- rating-update
- convergence-check
- cycle-detection
- ranking-synthesis
Execution Guidance
- Initialize all candidates with prior (mu=25, sigma=8.33 for TrueSkill)
- Run adaptive-pair-selection with uncertainty-based pair selection
- Prioritize comparisons that reduce uncertainty in top-k boundary
- Check convergence every N/10 comparisons
- When budget exhausted or converged, run ranking-synthesis
- Optional: spot-check consistency in top-10 region
Output Format
ranking:
- {rank: 1, candidate: "...", mu: 38.2, sigma: 1.4, ci: [35.4, 41.0]}
- {rank: 2, candidate: "...", mu: 36.8, sigma: 1.6, ci: [33.6, 40.0]}
method: trueskill
total_comparisons: 847
budget_utilization: 0.92
top_10_stability: 0.96
convergence_round: 14
Available Tactics
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use |
|---|---|
| adaptive-pair-selection | Iteratively select maximally informative pairs, execute comparisons, update ratings, and check convergence until ranking stabilizes. |
| consistency-audit-loop | Detect preference cycles, localize inconsistent judgments, request corrections, and recompute ratings until consistency threshold is met. |
Available SOPs
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use |
|---|---|
| ranking-synthesis | Produce the final ranking artifact from converged ratings and consistency report. |