メニュー
Search and pathfinding with A*, Q*, MCTS, minimax, BFS/DFS
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Test model robustness with adversarial attacks and defenses
Multi-armed bandit simulation — epsilon-greedy, UCB1, Thompson sampling
Benchmark and compare ix algorithm performance
Embedded Redis-like cache with TTL, LRU, pub/sub, and RESP protocol
Category theory primitives — monad laws verification, free-forgetful adjunction
Chaos theory analysis — Lyapunov exponents, bifurcation, attractors, fractals
| name | ix-search |
| description | Search and pathfinding with A*, Q*, MCTS, minimax, BFS/DFS |
| disable-model-invocation | true |
Find optimal paths, solve games, or explore state spaces.
When the user needs pathfinding, game AI, decision making under uncertainty, or combinatorial search.
| Problem | Algorithm | When |
|---|---|---|
| Shortest path with heuristic | A* | Known goal, admissible heuristic available |
| Learned heuristic search | Q* | When you can train a Q-function on the domain |
| Game tree (deterministic) | Minimax / Alpha-Beta | Two-player zero-sum games |
| Game tree (stochastic) | MCTS | Large branching factor, simulation possible |
| Unweighted shortest path | BFS | All edges cost 1 |
| Exhaustive exploration | DFS / IDDFS | Memory-constrained, solution depth unknown |
| Local optimization | Hill Climbing / Tabu | Single solution improvement |
use ix_search::astar::{SearchState, astar};
use ix_search::qstar::{QFunction, TabularQ, qstar_search};
use ix_search::mcts::{MctsState, mcts_search};
use ix_search::adversarial::{GameState, alpha_beta};
// Implement SearchState for your domain, then:
let result = astar(start_state, |s| s.manhattan_distance());
Q* uses a learned heuristic (DQN-style) instead of hand-crafted. Dramatically fewer node expansions in large action spaces. Use compare_qstar_vs_astar() to benchmark.