skills/symmetry-group-identifier/SKILL.md
Maps identified symmetries to mathematical groups (cyclic, dihedral, symmetric, SO(3), SE(3), E(3)) for equivariant neural network architecture design, using taxonomy and foundations from Visual Group Theory. Use when candidate symmetries have been identified and need formalization into group theory language, or when user mentions cyclic groups, dihedral groups, Lie groups, SO(3), SE(3), or permutation groups.
npx skillsauth add lyndonkl/claude symmetry-group-identifierInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Knowing your symmetry group tells you which equivariant architecture patterns to use. This skill formalizes identified transformations into the language of group theory.
Copy this checklist and track your progress:
Group Identification Progress:
- [ ] Step 1: List symmetries from discovery phase
- [ ] Step 2: Classify each as discrete or continuous
- [ ] Step 3: Match to specific groups using taxonomy
- [ ] Step 4: Determine how groups combine
- [ ] Step 5: Verify group properties
- [ ] Step 6: Document final group specification
Step 1: List symmetries from discovery phase
Gather the identified symmetries from the discovery phase. List each identified transformation and whether it requires invariance or equivariance. Note confidence levels. If symmetries haven't been discovered yet, work with user to identify them through domain analysis first.
Step 2: Classify each as discrete or continuous
For each symmetry, determine: Is the transformation set finite (discrete) or infinite (continuous)? Discrete examples: 90° rotations (4 elements), permutations of n items (n! elements). Continuous examples: rotation by any angle, translation by any distance. Use Group Taxonomy to guide classification. For mathematical foundations, see Group Theory Primer.
Step 3: Match to specific groups using taxonomy
Use the Discrete Groups and Continuous Groups reference sections. Identify the specific group name and notation for each symmetry. Common matches: n-fold rotation → Cₙ, rotation+reflection → Dₙ, permutation → Sₙ, 3D rotation → SO(3), rigid motion → SE(3), full Euclidean → E(3). For detailed Lie group information (SO(3), SE(3), E(3)), consult Lie Groups Reference.
Step 4: Determine how groups combine
If multiple symmetries are present, determine how they combine. Direct product (G × H): symmetries act independently. Semidirect product (G ⋊ H): one symmetry "twists" the other (e.g., SE(3) = SO(3) ⋊ ℝ³). Use Combining Groups reference.
Step 5: Verify group properties
Check that identified structure satisfies group axioms: closure, associativity, identity, inverses. Verify important properties: Is it compact? (affects representation theory). Is it abelian? (commutative or not). Is it connected? (affects implementation). Use Group Properties Checklist. For detailed verification methodology, see Methodology.
Step 6: Document final group specification
Create specification using Output Template. Include: group name/notation, dimension/size, key properties, invariance vs equivariance requirements, and recommended architecture family. This specification provides the foundation for architecture design. Quality criteria for this output are defined in Quality Rubric.
SYMMETRY GROUPS
│
┌───────────────┴───────────────┐
│ │
DISCRETE CONTINUOUS
│ (Lie Groups)
│ │
┌─────┼─────┐ ┌────────┼────────┐
│ │ │ │ │ │
Cyclic Dihedral Symmetric SO(n) SE(n) E(n)
Cₙ Dₙ Sₙ rotations rigid Euclidean
only motions (w/ reflect)
| Symmetry Type | Group | Notation | Elements | Common Use | |---------------|-------|----------|----------|------------| | n-fold rotation | Cyclic | Cₙ | n | Image rotation (90°, 60°) | | Rotation + reflection | Dihedral | Dₙ | 2n | Regular polygons | | Permutation | Symmetric | Sₙ | n! | Sets, graphs | | 2D rotation (continuous) | Special orthogonal | SO(2) | ∞ | Continuous rotation | | 3D rotation | Special orthogonal | SO(3) | ∞ | 3D orientation | | 3D rigid motion | Special Euclidean | SE(3) | ∞ | Robotics, molecules | | 3D with reflections | Euclidean | E(3) | ∞ | Chemistry, physics |
What they represent: Rotations by multiples of 360°/n
Elements: {e, r, r², ..., rⁿ⁻¹} where rⁿ = e (identity)
| Group | Rotations | Example | |-------|-----------|---------| | C₂ | 0°, 180° | Playing cards | | C₄ | 0°, 90°, 180°, 270° | Square images | | C₆ | 60° increments | Hexagonal patterns |
Use when: Rotation symmetry present but NOT reflection symmetry.
What they represent: Rotations + reflections of regular n-gon
Elements: n rotations + n reflections = 2n total
| Group | Elements | Example | |-------|----------|---------| | D₄ | 8 | Square with diagonals (p4m group) | | D₆ | 12 | Regular hexagon |
Use when: Both rotation AND reflection symmetry present.
What they represent: All permutations of n elements
Elements: n! permutations
Use when: Element ordering is arbitrary (sets, graphs, point clouds).
Elements: Rotation by any angle θ ∈ [0, 2π)
Matrix form: R(θ) = [[cos θ, -sin θ], [sin θ, cos θ]]
Use when: Continuous rotation symmetry in 2D.
Elements: All rotations in 3D (3 degrees of freedom)
Representations: Rotation matrices, quaternions, Euler angles, axis-angle
Use when: 3D orientation doesn't matter, but handedness does.
Elements: Rotations + Translations in 3D
Structure: SE(3) = SO(3) ⋊ ℝ³ (semidirect product)
Use when: Objects can be anywhere and in any orientation, handedness matters.
Elements: SE(3) + Reflections
Structure: E(3) = O(3) ⋊ ℝ³
Use when: SE(3) symmetry PLUS reflection symmetry (most molecules).
E(3) = O(3) ⋊ ℝ³
│ exclude reflections
▼
SE(3) = SO(3) ⋊ ℝ³
│ exclude translations
▼
SO(3)
│ 2D restriction
▼
SO(2)
When to use: Symmetries act independently (neither affects the other).
Example: Image with separate translation and color permutation → SE(2) × S₃
Property: (g₁, h₁) · (g₂, h₂) = (g₁g₂, h₁h₂)
When to use: One symmetry "twists" the other (don't commute).
Example: SE(3) = SO(3) ⋊ ℝ³ (rotating then translating ≠ translating then rotating)
Common cases: SE(n) = SO(n) ⋊ ℝⁿ, E(n) = O(n) ⋊ ℝⁿ, Dₙ = Cₙ ⋊ C₂
For your identified group, verify:
| Property | Question | Why It Matters | |----------|----------|----------------| | Compact | Is the group "bounded"? | Affects representation theory | | Abelian | Does order matter? (g₁g₂ = g₂g₁?) | Simplifies architecture | | Connected | Is group in one piece? | Affects irreducible representations | | Finite | Finite number of elements? | Discrete vs continuous architecture |
| Domain | Typical Group | Notes | |--------|--------------|-------| | 2D Image Classification | C₄ or D₄ | p4 or p4m groups | | 3D Molecular Energy | E(3) × Sₙ | Full Euclidean + atom permutation | | 3D Molecular Chirality | SE(3) × Sₙ | No reflections | | Point Cloud Classification | SO(3) × Sₙ | Rotation + permutation | | Graph Classification | Sₙ | Permutation invariant | | Robotics | SE(3) | Sometimes with gravity constraint |
SYMMETRY GROUP SPECIFICATION
============================
Identified Symmetries:
1. [Symmetry] → Group: [name] ([notation])
2. [Symmetry] → Group: [name] ([notation])
Combined Group Structure:
- Full group: [G₁ × G₂] or [G₁ ⋊ G₂]
- Size: [# elements] or [continuous]
Group Properties:
- Compact: [Yes/No]
- Abelian: [Yes/No]
- Connected: [Yes/No]
Symmetry Requirements:
- [Group]: [Invariant/Equivariant] for [task type]
Recommended Architecture Family:
- [Architecture] supporting [group]
NEXT STEPS:
- Empirically validate symmetry hypotheses if not yet confirmed
- Design equivariant architecture based on group specification
testing
Cluster a conference's event records into a small set of coarse themes with finer sub-clusters, an explicit outlier bucket, and soft (multi-membership) affinities — using the hybrid embed-then-label pipeline (embed abstracts, reduce, density-cluster, then LLM-label the clusters) when embedding libraries are available, and an LLM-reasoned hierarchical fallback when they are not. Embeddings do the grouping; the LLM only names the groups. Conference-agnostic. Use when turning structured event records into a navigable theme map for preference elicitation and scheduling, when you need 6-8 reasonable themes rather than 20 muddy ones, or when overlapping talks must belong to more than one theme. Trigger keywords - theme clustering, cluster talks, embed then label, soft membership, outlier talks, conference themes, topic map.
development
Build a personal conference schedule as a constraint-optimization problem — hard constraints (no time overlap, room-to-room travel time, capacity/registration, the attendee's own must-attends and blackouts) plus a user-owned weighted objective trading interest against breadth, pacing (maximize contiguous free time), and serendipity. Surfaces unbreakable conflicts (two high-value overlapping talks the model cannot rank) as decisions for the human rather than silently picking, and reports what each choice traded away. Conference-agnostic. Use to turn a preference profile plus a theme map into a day-by-day plan, to resolve overlapping sessions, or to balance a packed vs paced schedule. Trigger keywords - schedule optimization, conference schedule, constraint optimization, overlapping talks, contiguous free time, conflict surfacing, packed vs paced.
development
Parse a heterogeneous conference program (markdown, HTML, PDF-derived text, or JSON) into normalized event records with per-field confidence scores and independent classification axes (topic, depth, format, prerequisites, recorded, capacity). Detects the program's format before extracting, treats every inferred field as uncertain (present vs inferred vs missing), and flags thin or missing abstracts so downstream enrichment can target them. Conference-agnostic. Use when ingesting a conference or event schedule into a structured store, normalizing a talk/session list, or extracting per-session metadata with calibrated confidence. Trigger keywords - program ingestion, parse schedule, session extraction, event records, conference program, talk metadata, per-field confidence.
development
Build a personalized preference profile from a small number of well-chosen, cluster-grounded questions instead of a long survey. Represents the person's interests as an uncertainty region over the theme map, picks the single highest-information-gain choice-based question (contrasting real talks from different clusters), balances exploiting known interests against exploring uncertain ones, deliberately injects outlier probes to fight selection bias, and stops as soon as the schedule would be stable. Also elicits the user-owned objective weights and hard constraints. Interactive — runs where it can actually ask the person. Conference-agnostic. Use to turn a theme map into a preference profile, to decide what to ask a conference attendee, or to elicit scheduling priorities. Trigger keywords - preference elicitation, ask few questions, information gain, choice-based questions, selection bias probe, objective weights, attendee preferences.