Эксперимент: Canonical orbit search for symmetric pruning masks (#1307)

{ "worked": true, "confidence": 9, "verdict": "Built an exact permutation-group canonicalizer, canonical DFS, orbit coverage checks, and a two-block symmetry experiment. The math checks passed: canonical representatives covered every mask orbit, DFS matched the canonical set, predecessor closure held, and the lexicographic order check passed. Under full exchangeability, 70 masks collapsed to 1; under two exchangeable blocks, 70 masks collapsed to 5 with identical best score and 14x fewer evaluated candidates. The reduction is real, but naive factorial canonicality testing was slower in wall-clock time than exhaustive scanning at this tiny size.", "metrics": { "baseline": "70 masks; two-block best score 12; exhaustive scan 0.000064 s", "idea": "5 canonical representatives; best score 12; 14x evaluation reduction; canonical DFS 0.0264 s", "toy_verification": "n=7,k=3: 35 masks, 1 orbit, DFS/reps equal, orbit coverage true, predecessor closure true" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 canonical_orbit.py && /home/maxwelhelp/main/bin/python3 experiment.py", "files": [ "canonical_orbit.py", "experiment.py" ], "limitations": "No neural-network training or real pruning accuracy experiment was run. The implementation enumerates explicit permutation groups, so factorial group construction and canonicality scans do not scale; a stabilizer-chain or graph-canonical-labeling backend is needed for larger channel counts." }