lingzhi227/agent-research-skills

symbolic-equation

Discover scientific equations from data using LLM-guided evolutionary search (LLM-SR).

ソースを見る

使い方

  1. インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
  2. 必要なツール、認証情報、対応環境を原文で確認します。
  3. 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/lingzhi227/agent-research-skills --skill symbolic-equation

開始前の確認

  • 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
  • スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
リポジトリの原文

見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。

Symbolic Equation Discovery

Discover interpretable scientific equations from data using LLM-guided evolutionary search.

Input

  • $0 — Dataset description, variable names, and physical context

References

  • LLM-SR patterns (prompts, evolution, sampling): ~/.claude/skills/symbolic-equation/references/llmsr-patterns.md

Workflow (from LLM-SR)

Step 1: Define Problem Specification

Create a specification with:

  1. Input variables: Physical quantities with types (e.g., x: np.ndarray, v: np.ndarray)
  2. Output variable: Target quantity to predict
  3. Evaluation function: Fitness metric (typically negative MSE with parameter optimization)
  4. Physical context: Domain knowledge to guide equation discovery
python
# Example specification
@equation.evolve
def equation(x: np.ndarray, v: np.ndarray, params: np.ndarray) -> np.ndarray:
    """Describe the acceleration of a damped nonlinear oscillator."""
    return params[0] * x

Step 2: Initialize Multi-Island Buffer

  • Create N islands (default: 10) for population diversity
  • Each island maintains independent clusters of equations
  • Clusters group equations by performance signature

Step 3: Evolutionary Search Loop

Repeat until convergence or max samples:

  1. Select island: Random island selection
  2. Build prompt: Sample top equations from clusters (softmax-weighted by score)
  3. LLM proposes: Generate new equation as improved version
  4. Evaluate: Execute on test data, compute fitness score
  5. Register: Add to island's cluster if valid

Step 4: Prompt Construction

Present previous equations as versioned sequence:

python
def equation_v0(x, v, params):
    """Initial version."""
    return params[0] * x

def equation_v1(x, v, params):
    """Improved version of equation_v0."""
    return params[0] * x + params[1] * v

def equation_v2(x, v, params):
    """Improved version of equation_v1."""
    # LLM completes this

Step 5: Island Reset (Diversity Maintenance)

Periodically (default: every 4 hours):

  1. Sort islands by best score
  2. Reset bottom 50% of islands
  3. Seed each reset island with best equation from a surviving island
  4. Restart cluster sampling temperature

Step 6: Extract Best Equations

After search completes:

  1. Collect best equation from each island
  2. Rank by fitness score
  3. Simplify if possible (algebraic simplification)
  4. Report with physical interpretation

Cluster Sampling

Temperature-scheduled softmax over cluster scores:

temperature = T_init * (1 - (num_programs % period) / period)
probabilities = softmax(cluster_scores / temperature)
  • Higher temperature → more exploration
  • Lower temperature → more exploitation of best clusters
  • Within clusters: shorter programs are preferred (Occam's razor)

Rules

  • Equations must use only standard mathematical operations
  • Parameter optimization via scipy BFGS or Adam
  • Fitness = negative MSE (higher is better)
  • Timeout protection for equation evaluation
  • No recursive equations allowed
  • Physical interpretability is preferred over pure fit

Related Skills

タスクに関連する項目

同じリポジトリから

関連する Skills

すべての Skills
lingzhi227
コミュニティ

backward-traceability

Make every number in the final PDF traceable to the exact code line that produced it. Uses \hypertarget/\hyperlink LaTeX commands and \num{formula} evaluated at compile time. Use for reproducibility and data integrity verification.

1時間の導入数
2
GitHub Stars
451
更新日
2月27日
lingzhi227
コミュニティ

experiment-code

Write ML experiment code with iterative improvement. Generate training/evaluation pipelines, debug errors, and optimize results through code reflection. Use when implementing experiments for a research paper.

1時間の導入数
1
GitHub Stars
420
更新日
2月27日
lingzhi227
コミュニティ

paper-assembly

Orchestrate the full paper pipeline end-to-end. Manage state propagation between phases (literature → plan → code → experiments → figures → tables → writing → review), support checkpointing and resumption. Use for assembling a complete paper from components.

1時間の導入数
1
GitHub Stars
420
更新日
2月27日
lingzhi227
コミュニティ

literature-review

Conduct comprehensive literature reviews using multi-perspective dialogue simulation. Generate diverse expert personas, conduct grounded Q&A conversations, and synthesize findings into structured knowledge. Use when starting a new research project or writing a survey section.

1時間の導入数
1
GitHub Stars
414
更新日
2月27日