symmetry-discovery-questionnaire
$
npx mdskill add lyndonkl/claude/symmetry-discovery-questionnaireGuide ML engineers to uncover hidden data symmetries without group theory.
- Identifies invariance patterns when users discuss symmetry or model constraints.
- Classifies domains to suggest transformations like rotation or permutation.
- Executes a six-step checklist from domain analysis to documentation.
- Delivers a structured progress tracker with actionable domain-specific steps.
SKILL.md
.github/skills/symmetry-discovery-questionnaireView on GitHub ↗
--- name: symmetry-discovery-questionnaire description: Guides collaborative discovery of hidden symmetries in ML data through structured domain analysis, coordinate system examination, transformation testing, and physical constraint identification. No group theory knowledge required. Use when ML engineers need to identify symmetries in their data, when user mentions data symmetry, invariance discovery, what transformations matter, or needs help recognizing patterns their model should respect. --- # Symmetry Discovery Questionnaire ## Workflow Copy this checklist and track your progress: ``` Symmetry Discovery Progress: - [ ] Step 1: Classify your domain and data type - [ ] Step 2: Analyze coordinate system choices - [ ] Step 3: Test candidate transformations - [ ] Step 4: Analyze physical constraints - [ ] Step 5: Determine output behavior under transformations - [ ] Step 6: Document symmetry candidates ``` **Step 1: Classify your domain and data type** Ask user what their primary data type is. Use this table to identify likely symmetries and guide further questions. Images (2D grids) → likely translation, rotation, reflection. 3D data (point clouds, meshes) → likely SE(3), E(3). Molecules → E(3) + permutation + point groups. Graphs/Networks → permutation. Sets → permutation. Time series → time-translation, periodicity. Tabular → rarely symmetric. Physical systems → conservation laws imply symmetries. For detailed worked examples by domain, consult [Domain Examples](./resources/domain-examples.md). **Step 2: Analyze coordinate system choices** Guide user through coordinate analysis questions: Is there a preferred origin? (NO → translation invariance). Is there a preferred orientation? (NO → rotation invariance). Is there a preferred handedness? (NO → reflection invariance). Is there a preferred scale? (NO → scale invariance). Is element ordering meaningful? (NO → permutation invariance). Document each answer with reasoning. **Step 3: Test candidate transformations** For each candidate transformation T, ask: "If I transform my input by T, should my output change?" If NO → invariance to T. If YES predictably → equivariance to T. If YES unpredictably → no symmetry. Use domain-specific checklists from [Domain Transformation Tests](#domain-transformation-tests). Test all relevant transformations systematically. For the detailed methodology behind this testing approach, see [Methodology](./resources/methodology.md). **Step 4: Analyze physical constraints** Ask about conservation laws and physical symmetries. Noether's theorem: every conservation law implies a symmetry. Energy conserved → time-translation symmetry. Momentum conserved → space-translation symmetry. Angular momentum conserved → rotation symmetry. Ask: Are there physical conservation laws? Is system isolated from external reference frames? Are there gauge freedoms? **Step 5: Determine output behavior under transformations** Critical question: When input transforms, how should output transform? Classification labels → stay same (invariance). Bounding boxes → move with object (equivariance). Force vectors → rotate with system (equivariance). Scalar properties → stay same (invariance). Segmentation masks → transform with image (equivariance). This determines whether you need invariant or equivariant architecture. **Step 6: Document symmetry candidates** Create summary using [Output Template](#output-template). List identified symmetries with confidence levels. Note uncertain cases that need empirical validation. Identify non-symmetries (transformations that DO matter). Recommend next steps for validation and formalization. Quality criteria for this output are defined in [Quality Rubric](./resources/evaluators/rubric_symmetry_discovery.json). ## Domain Transformation Tests ### Image Symmetries | Transformation | Test Question | If NO → | |----------------|---------------|---------| | Translation | Does object position matter for label? | Translation invariance | | Rotation (90°) | Would rotated image have same label? | C4 symmetry | | Rotation (any) | Would any rotation preserve label? | SO(2) symmetry | | Horizontal flip | Would mirror image have same label? | Reflection | | Scale | Would zoomed image have same label? | Scale invariance | ### 3D Data Symmetries | Transformation | Test Question | If NO → | |----------------|---------------|---------| | 3D Translation | Does absolute position matter? | Translation invariance | | 3D Rotation | Does orientation matter? | SO(3) or SE(3) | | Reflection | Does handedness matter? | O(3) or E(3) | | Point permutation | Does point ordering matter? | Permutation invariance | ### Graph Symmetries | Transformation | Test Question | If NO → | |----------------|---------------|---------| | Node relabeling | Does node ID matter, or just connectivity? | Permutation invariance | ### Molecular Symmetries | Transformation | Test Question | If NO → | |----------------|---------------|---------| | Rotation | Is property independent of orientation? | SO(3) | | Translation | Is property independent of position? | Translation | | Reflection | Are both enantiomers equivalent? | Include reflections | | Atom permutation | Do identical atoms behave identically? | Permutation | ### Temporal Symmetries | Transformation | Test Question | If NO → | |----------------|---------------|---------| | Time shift | Can pattern occur at any time? | Time-translation | | Time reversal | Is forward same as backward? | Time-reversal | | Periodicity | Do patterns repeat with period T? | Cyclic symmetry | ## Quick Reference **The 5 Key Questions:** 1. Is there a preferred coordinate system? (origin, orientation, scale) 2. Does element ordering matter? 3. What transformations leave the label unchanged? 4. What physical constraints apply? 5. How should outputs transform when inputs transform? **Common Symmetry → Group Mapping:** - Rotation (2D, discrete) → Cyclic group Cₙ - Rotation + reflection (2D) → Dihedral group Dₙ - Rotation (2D, continuous) → SO(2) - Rotation (3D) → SO(3) - Rotation + translation (3D) → SE(3) - Full Euclidean (3D) → E(3) - Permutation → Symmetric group Sₙ ## Output Template ``` SYMMETRY CANDIDATE SUMMARY ========================== Domain: [Data type] Task: [Classification/Regression/Detection/etc.] IDENTIFIED SYMMETRIES: 1. [Transformation]: [Invariance/Equivariance] - Evidence: [Why you believe this] - Confidence: [High/Medium/Low] 2. [Transformation]: [Invariance/Equivariance] - Evidence: [Why you believe this] - Confidence: [High/Medium/Low] UNCERTAIN SYMMETRIES (need validation): - [Transformation]: [Reason for uncertainty] NON-SYMMETRIES (transformations that DO matter): - [Transformation]: [Why it matters] NEXT STEPS: - Empirically validate uncertain symmetry candidates - Map confirmed symmetries to mathematical groups - Design architecture based on validated group structure ```
More from lyndonkl/claude
- abstraction-concrete-examplesBuilds structured abstraction ladders that translate high-level principles into concrete, actionable examples across 3-5 levels. Bridges communication gaps, reveals hidden assumptions, and tests whether abstract ideas work in practice. Use when explaining concepts at different expertise levels, moving between abstract principles and concrete implementation, identifying edge cases by testing ideas against scenarios, designing layered documentation, decomposing complex problems into actionable steps, or bridging strategy-execution gaps.
- academic-letter-architectGuides the creation of evidence-based academic recommendation letters, reference letters, and award nominations that combine concrete examples, meaningful comparisons, and genuine enthusiasm. Use when writing recommendation letters for students, postdocs, or colleagues, or when user mentions recommendation letter, reference, nomination, letter of support, endorsement, or needs help with strong advocacy and comparative statements.
- adr-architectureDocuments significant architectural and technical decisions with full context, alternatives considered, trade-offs analyzed, and consequences understood. Creates a decision trail that helps teams understand why decisions were made. Use when choosing between technology options, making infrastructure decisions, establishing standards, migrating systems, or when user mentions ADR, architecture decision, technical decision record, or decision documentation.
- adverse-selection-priorProduces a Bayesian prior probability that an offered transaction is +EV for the recipient, given that the counterparty chose to propose it. Applies Akerlof market-for-lemons logic -- if they offered it, they believe it is +EV for them, so the prior that it is +EV for us is materially below 50%. Reusable across trade evaluation, waiver drops (another team dropping a player is also adverse selection), job-offer analysis, M&A, and any "someone offered me this" situation. Use when you receive an unsolicited trade/offer/proposal, analyzing incoming trade prior, evaluating why a counterparty proposed a deal, or when user mentions adverse selection, market for lemons, why did they offer this, incoming trade prior, they proposed it, Bayesian adjustment on received offer.
- alignment-values-north-starCreates actionable alignment frameworks that give teams a shared North Star (direction), values (guardrails), and decision tenets (behavioral standards). Enables autonomous decision-making while maintaining organizational coherence. Use when starting new teams, scaling organizations, defining culture, establishing product vision, resolving misalignment, creating strategic clarity, or when user mentions North Star, team values, mission, principles, guardrails, decision framework, or cultural alignment.
- analogy-weight-checkFor every analogy in a substacker draft, verifies it carries mechanical weight — the analogy does real work explaining the mechanism, not merely decorates it. Cross-references analogy-catalog.md for novelty (is this analogy reused from a prior post?) and domain fit (biology > organizational > sports preferred; physics/military disfavored). Use whenever an analogy appears in the draft. Trigger keywords: analogy weight, decorative, mechanical weight, reused analogy, catalog check, metaphor check.
- answer-uncomfortable-questionTakes one strategic question about substacker ("should we launch paid?", "is this section dead?", "are we writing for the wrong audience?") and produces the mandatory evidence + reasoning + downside triad plus a recommendation. Used 3 times per Growth Strategist review. Trigger keywords: uncomfortable question, strategic question, evidence reasoning downside, triad.
- attribute-performanceFor each substacker post that materially over- or under-performs the rolling baseline (|z| ≥ 1.0), produces a plain-English attribution paragraph with calibrated confidence (high / medium / low / unexplained). Considers subject-line effect, topic zeitgeist, external share, day-of-week, length effect, and audience-notes signals. Labels unexplained outliers explicitly rather than fabricating a story. Use after compute-baseline when outlier posts exist. Trigger keywords: attribution, why did this post work, outlier explanation, performance analysis.
- auction-first-price-shadingComputes the optimal shaded bid for a first-price sealed-bid auction given a true private value, an estimate of the number of competing bidders N, and a value-distribution assumption. Implements the `(N-1)/N` equilibrium shading rule for uniform private values, adjusts for log-normal or empirical value distributions, layers a risk-aversion adjustment, and caps output against the bidder's remaining budget. Domain-neutral auction theory reusable across fantasy sports (baseball FAAB, NBA/NHL waiver auctions), prediction-market limit sizing, sealed procurement bids, and any blind-bid context. Use when user mentions "first-price auction bid", "sealed bid shading", "(N-1)/N", "FAAB bid amount", "auction shading", "optimal bid first-price", "bid for sealed-bid", "blind bid sizing", or when downstream logic needs a principled shade factor rather than an ad-hoc heuristic.
- auction-winners-curse-haircutApplies a Bayesian haircut to a bid valuation for common-value auctions where winning is itself evidence the bidder over-estimated. Takes a raw valuation, a value-type classification (common_value / private_value / mixed), the number of informed bidders N, and a signal-dispersion estimate, and returns an adjusted valuation. Domain-neutral and reusable across fantasy FAAB, prediction markets, M&A bids, ad-auction budgets, and any generic bidding context. Use when user mentions "winner's curse", "common value auction", "valuation haircut", "adverse valuation", "Bayesian bid adjustment", or "over-paying in auction".