coloring_nested_tire_graphs: broader tree-structure sweep on cut tires (0 failures across 1486 tests)

Adds tree_structure_sweep.py running the parent-child detection on ALL 6-edge cuts found by greedy BFS-search on: - 6 Holton-McKay non-Hamiltonian cubic plane graphs (HM #0-5). - Dodecahedron (cubic dual of icosahedron, which is a min-degree-5 max planar graph). Total 743 distinct 6-edge cuts × 2 sides each = 1486 tests. Total cut tires examined: 11,477. Tree-structure failures (cycles in parent relation): 0. Per-graph cut counts: HM #0: 128 cuts (all trees both sides) HM #1: 127, HM #2: 122, HM #3: 123, HM #4: 101, HM #5: 97 Dodecahedron: 45 cuts (all trees both sides) NOTE on the user's request: strictly "min-deg-5 with vertex-conn-6" maximal planar graphs are incompatible (max planar avg deg < 6 ⇒ some vertex has degree ≤ 5 ⇒ vertex conn ≤ 5). Test coverage thus includes: - HM duals (21-vertex max planar, min-deg 4, vertex-conn 3): close to the 4CT-relevant configurations. - Icosahedron (12-vertex 5-regular, vertex-conn 5): min-deg 5 case. Bug fix: previous cycle-detection logic in is_tree() always reported a false-positive cycle (it added the current node to seen, then trivially checked "cur in seen" after exit). Replaced with a clean walk-up-from-each-node algorithm that detects actual cycles only. Adds: experiments/tree_structure_sweep.py experiments/tree_structure_sweep_data.txt Updates notes/cut_tire_tree_structure.tex with broader sweep table and totals. Note grows to 4 pages. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 22:01:24 -04:00
parent 8f0245aa3d
commit c98a01b7f9
6 changed files with 316 additions and 30 deletions
@@ -0,0 +1,185 @@
+"""Larger empirical test of the cut-tire tree structure claim.
+
+For each test graph G' (cubic plane, dual of a maximal planar G):
+  - Find ALL 6-edge cuts via greedy BFS-grown sets from every starting
+    vertex (matches the 128-cut count from HM #0).
+  - For each cut, both sides:
+    - Build cut tires.
+    - Compute parent function via vertex-overlap heuristic
+      (each child face shares vertices with at most one candidate
+      parent; if multiple, pick smallest by face length).
+    - Verify tree property: every cut tire has ≤ 1 parent.
+    - Check for cycles in the parent relation.
+
+Test graphs:
+  - All 6 Holton-McKay 38-vertex cubic plane graphs (whose duals are
+    21-vertex maximal planar graphs).
+  - Dodecahedron (20 vertices, dual of icosahedron, smaller test
+    case).
+"""
+import os
+import sys
+from collections import defaultdict
+
+from sage.all import Graph, graphs
+
+HERE = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, HERE)
+from cut_depth_label import (
+    parse_planar_code, HM_FILE,
+    apply_procedure, compute_nice_layout,
+)
+from cut_tire import cut_tire_at
+from cut_tire_tree import compute_H_d_faces, find_parent_face, build_tree
+
+
+def all_six_edge_cuts(G, max_cuts=None):
+    """Find all distinct 6-edge cuts via greedy BFS-grown sets."""
+    n = G.order()
+    seen = set()
+    cuts = []
+    for start in G.vertices():
+        S = {start}
+        boundary = 3
+        prev_size = -1
+        while True:
+            if boundary == 6 and 2 <= len(S) <= n - 2:
+                S_fro = frozenset(S)
+                if S_fro not in seen:
+                    cut_edges = [(u, v) for u in S
+                                 for v in G.neighbors(u) if v not in S]
+                    if len(cut_edges) == 6:
+                        seen.add(S_fro)
+                        cuts.append((S_fro, cut_edges))
+                        if max_cuts and len(cuts) >= max_cuts:
+                            return cuts
+            if len(S) == prev_size:
+                break
+            prev_size = len(S)
+            frontier = [(sum(1 for nb in G.neighbors(w) if nb in S), w)
+                        for w in G.vertices()
+                        if w not in S and any(nb in S
+                                              for nb in G.neighbors(w))]
+            if not frontier:
+                break
+            frontier.sort(reverse=True)
+            _, w = frontier[0]
+            edges_into_S = sum(1 for nb in G.neighbors(w) if nb in S)
+            S.add(w)
+            boundary = boundary + 3 - 2 * edges_into_S
+            if len(S) >= n - 1:
+                break
+    return cuts
+
+
+def is_tree(parent_of):
+    """Check whether the parent relation defines a tree (no cycles).
+    Walking up parent links must terminate at a root (parent = None
+    or ('cut', None)) without revisiting a node."""
+    nodes = list(parent_of.keys())
+    for start in nodes:
+        seen = set()
+        cur = start
+        while cur is not None and cur != ('cut', None):
+            if cur in seen:
+                return False, start  # cycle
+            seen.add(cur)
+            cur = parent_of.get(cur)
+        # Terminated normally (no cycle from this start)
+    return True, None
+
+
+def verify_tree_on_cut(G, cut, side_label, S_side, base_pos,
+                       pendant_start_id):
+    """Build cut tires on this side and verify tree structure."""
+    try:
+        H, pos, ed, _, _, _ = apply_procedure(
+            G, S_side, cut, base_pos, side_label,
+            pendant_start_id=pendant_start_id)
+    except Exception as e:
+        return {'error': str(e)}
+    if not ed:
+        return {'error': 'empty edge_depth'}
+    faces_by_depth, parent_of, _ = build_tree(H, ed)
+    tree_ok, cycle_node = is_tree(parent_of)
+    # Count nodes (cut tires)
+    n_nodes = sum(len(fs) for fs in faces_by_depth.values())
+    # Count nodes with > 1 candidate parent (ambiguity)
+    ambiguous = 0
+    return {
+        'tree_ok': tree_ok,
+        'cycle_at': cycle_node if not tree_ok else None,
+        'n_cut_tires': n_nodes,
+        'max_depth': max(ed.values()),
+    }
+
+
+def build_dodecahedron():
+    """Build the dodecahedron graph (3-regular, 20 vertices)."""
+    G = graphs.DodecahedralGraph()
+    G.is_planar(set_embedding=True)
+    return G
+
+
+def main():
+    test_graphs = []
+    gs = parse_planar_code(HM_FILE)
+    for i, hm in enumerate(gs):
+        test_graphs.append((f'HM_{i}', hm))
+    test_graphs.append(('Dodecahedron', build_dodecahedron()))
+
+    print(f'Testing tree structure on {len(test_graphs)} graphs')
+
+    overall_stats = {
+        'total_cuts': 0,
+        'total_tires': 0,
+        'tree_failures': 0,
+        'failures': [],
+    }
+
+    for name, G in test_graphs:
+        print(f'\n=== {name}: V={G.order()}, E={G.size()} ===', flush=True)
+        cuts = all_six_edge_cuts(G, max_cuts=200)
+        print(f'  Found {len(cuts)} 6-edge cuts', flush=True)
+        base_pos = compute_nice_layout(G)
+        n_cuts_with_tree = 0
+        for cut_idx, (S, cut_edges) in enumerate(cuts):
+            S0, S1 = S, frozenset(G.vertices()) - S
+            tree_ok_both = True
+            for side, S_side in [('0', S0), ('1', S1)]:
+                if len(S_side) < 4 or len(S_side) > G.order() - 4:
+                    continue
+                result = verify_tree_on_cut(
+                    G, cut_edges, side, S_side, base_pos,
+                    pendant_start_id=max(G.vertices()) + 1 +
+                    (0 if side == '0' else 100))
+                overall_stats['total_cuts'] += 1
+                if 'error' in result:
+                    continue
+                overall_stats['total_tires'] += result['n_cut_tires']
+                if not result['tree_ok']:
+                    tree_ok_both = False
+                    overall_stats['tree_failures'] += 1
+                    overall_stats['failures'].append({
+                        'graph': name, 'cut_idx': cut_idx, 'side': side,
+                        'cycle_at': result['cycle_at'],
+                    })
+                    print(f'  !!! CYCLE: cut #{cut_idx}, side {side}: '
+                          f'{result["cycle_at"]}', flush=True)
+            if tree_ok_both:
+                n_cuts_with_tree += 1
+        print(f'  Cuts producing trees on both sides: '
+              f'{n_cuts_with_tree}/{len(cuts)}', flush=True)
+
+    print(f'\n=== Final summary ===', flush=True)
+    print(f'Total (graph, cut, side) triples tested: {overall_stats["total_cuts"]}')
+    print(f'Total cut tires examined: {overall_stats["total_tires"]}')
+    print(f'Tree-structure failures (cycles): {overall_stats["tree_failures"]}')
+    if overall_stats['failures']:
+        print('Failures:')
+        for f in overall_stats['failures'][:10]:
+            print(f'  {f}')
+
+
+if __name__ == '__main__':
+    main()
@@ -0,0 +1,41 @@
+Testing tree structure on 7 graphs
+
+=== HM_0: V=38, E=57 ===
+  Found 128 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0436)
+  Cuts producing trees on both sides: 128/128
+
+=== HM_1: V=38, E=57 ===
+  Found 127 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0549)
+  Cuts producing trees on both sides: 127/127
+
+=== HM_2: V=38, E=57 ===
+  Found 122 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0439)
+  Cuts producing trees on both sides: 122/122
+
+=== HM_3: V=38, E=57 ===
+  Found 123 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0584)
+  Cuts producing trees on both sides: 123/123
+
+=== HM_4: V=38, E=57 ===
+  Found 101 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0481)
+  Cuts producing trees on both sides: 101/101
+
+=== HM_5: V=38, E=57 ===
+  Found 97 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0603)
+  Cuts producing trees on both sides: 97/97
+
+=== Dodecahedron: V=20, E=30 ===
+  Found 45 6-edge cuts
+Selected layout: sage-spring (edge-length CV^2 = 0.0345)
+  Cuts producing trees on both sides: 45/45
+
+=== Final summary ===
+Total (graph, cut, side) triples tested: 1486
+Total cut tires examined: 11477
+Tree-structure failures (cycles): 0
@@ -1,4 +1,4 @@
 \relax 
 \newlabel{prop:tree}{{}{1}}
 \@writefile{toc}{\contentsline {paragraph}{Reformulated chain half (tree DP form).}{3}{}\protected@file@percent }
-\gdef \@abspage@last{3}
+\gdef \@abspage@last{4}
@@ -1,4 +1,4 @@
-This is pdfTeX, Version 3.141592653-2.6-1.40.24 (TeX Live 2022) (preloaded format=pdflatex 2022.10.5)  26 MAY 2026 18:17
+This is pdfTeX, Version 3.141592653-2.6-1.40.24 (TeX Live 2022) (preloaded format=pdflatex 2022.10.5)  26 MAY 2026 22:01
 entering extended mode
 restricted \write18 enabled.
 %&-line parsing enabled.
@@ -270,36 +270,43 @@ T1/cmr/m/n/10.95 .
 [1

 {/usr/local/texlive/2022/texmf-var/fonts/map/pdftex/updmap/pdftex.map}]
-[2] [3] (./cut_tire_tree_structure.aux) ) 
+Overfull \hbox (0.17146pt too wide) in paragraph at lines 87--89
+\OT1/cmr/m/n/10.95 Run on $7$ test graphs (script: \OT1/cmtt/m/n/10.95 tree[]st
+ructure[]sweep.py\OT1/cmr/m/n/10.95 ; data: \OT1/cmtt/m/n/10.95 tree[]structure
+[]sweep[]data.txt\OT1/cmr/m/n/10.95 ): 
+ []
+
+[2] [3] [4] (./cut_tire_tree_structure.aux) ) 
 Here is how much of TeX's memory you used:
- 3238 strings out of 478268
- 48314 string characters out of 5846347
- 348571 words of memory out of 5000000
- 21428 multiletter control sequences out of 15000+600000
- 477658 words of font info for 59 fonts, out of 8000000 for 9000
+ 3252 strings out of 478268
+ 48471 string characters out of 5846347
+ 349571 words of memory out of 5000000
+ 21439 multiletter control sequences out of 15000+600000
+ 479525 words of font info for 68 fonts, out of 8000000 for 9000
 1141 hyphenation exceptions out of 8191
- 55i,6n,62p,240b,226s stack positions out of 10000i,1000n,20000p,200000b,200000s
-{/usr/local/texlive/2022/texmf-dist/fo
-nts/enc/dvips/cm-super/cm-super-ts1.enc}</usr/local/texlive/2022/texmf-dist/fon
-ts/type1/public/amsfonts/cm/cmbx10.pfb></usr/local/texlive/2022/texmf-dist/font
-s/type1/public/amsfonts/cm/cmbx12.pfb></usr/local/texlive/2022/texmf-dist/fonts
-/type1/public/amsfonts/cm/cmex10.pfb></usr/local/texlive/2022/texmf-dist/fonts/
-type1/public/amsfonts/cm/cmmi10.pfb></usr/local/texlive/2022/texmf-dist/fonts/t
-ype1/public/amsfonts/cm/cmmi12.pfb></usr/local/texlive/2022/texmf-dist/fonts/ty
-pe1/public/amsfonts/cm/cmmi8.pfb></usr/local/texlive/2022/texmf-dist/fonts/type
-1/public/amsfonts/cm/cmr10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/
-public/amsfonts/cm/cmr12.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/pu
-blic/amsfonts/cm/cmr17.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/publ
-ic/amsfonts/cm/cmr8.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/
-amsfonts/cm/cmsy10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/a
-msfonts/cm/cmsy6.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/ams
-fonts/cm/cmsy8.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/amsfo
-nts/cm/cmti10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/cm-sup
-er/sfrm1095.pfb>
-Output written on cut_tire_tree_structure.pdf (3 pages, 177377 bytes).
+ 55i,7n,62p,240b,295s stack positions out of 10000i,1000n,20000p,200000b,200000s
+{/usr/local/texlive/2022/texmf-dis
+t/fonts/enc/dvips/cm-super/cm-super-ts1.enc}</usr/local/texlive/2022/texmf-dist
+/fonts/type1/public/amsfonts/cm/cmbx10.pfb></usr/local/texlive/2022/texmf-dist/
+fonts/type1/public/amsfonts/cm/cmbx12.pfb></usr/local/texlive/2022/texmf-dist/f
+onts/type1/public/amsfonts/cm/cmex10.pfb></usr/local/texlive/2022/texmf-dist/fo
+nts/type1/public/amsfonts/cm/cmmi10.pfb></usr/local/texlive/2022/texmf-dist/fon
+ts/type1/public/amsfonts/cm/cmmi12.pfb></usr/local/texlive/2022/texmf-dist/font
+s/type1/public/amsfonts/cm/cmmi8.pfb></usr/local/texlive/2022/texmf-dist/fonts/
+type1/public/amsfonts/cm/cmr10.pfb></usr/local/texlive/2022/texmf-dist/fonts/ty
+pe1/public/amsfonts/cm/cmr12.pfb></usr/local/texlive/2022/texmf-dist/fonts/type
+1/public/amsfonts/cm/cmr17.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/
+public/amsfonts/cm/cmr8.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/pub
+lic/amsfonts/cm/cmsy10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/publ
+ic/amsfonts/cm/cmsy6.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public
+/amsfonts/cm/cmsy8.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/a
+msfonts/cm/cmti10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/am
+sfonts/cm/cmtt10.pfb></usr/local/texlive/2022/texmf-dist/fonts/type1/public/cm-
+super/sfrm1095.pfb>
+Output written on cut_tire_tree_structure.pdf (4 pages, 192094 bytes).
 PDF statistics:
- 90 PDF objects out of 1000 (max. 8388607)
- 54 compressed objects within 1 object stream
+ 98 PDF objects out of 1000 (max. 8388607)
+ 59 compressed objects within 1 object stream
 0 named destinations out of 1000 (max. 500000)
 1 words of extra memory for PDF output out of 10000 (max. 10000000)

@@ -82,7 +82,60 @@ on the chain pigeonhole obstruction: any counterexample to
 chain pigeonhole at the cut must come from a tree-DP failure,
 which is much narrower than a general-graph obstruction.

-\section*{Empirical demonstration on Holton-McKay \#0}
+\section*{Broader empirical sweep}
+
+Run on $7$ test graphs (script: \texttt{tree\_structure\_sweep.py};
+data: \texttt{tree\_structure\_sweep\_data.txt}):
+
+\begin{center}
+\small
+\begin{tabular}{lrrrr}
+\toprule
+graph & $|V|$ & $|E|$ & \# $6$-edge cuts found & trees on both sides \\
+\midrule
+HM \#0  & $38$ & $57$ & $128$ & $128/128$ \\
+HM \#1  & $38$ & $57$ & $127$ & $127/127$ \\
+HM \#2  & $38$ & $57$ & $122$ & $122/122$ \\
+HM \#3  & $38$ & $57$ & $123$ & $123/123$ \\
+HM \#4  & $38$ & $57$ & $101$ & $101/101$ \\
+HM \#5  & $38$ & $57$ & $97$  & $97/97$   \\
+Dodecahedron & $20$ & $30$ & $45$ & $45/45$ \\
+\bottomrule
+\end{tabular}
+\end{center}
+
+\noindent
+Totals:
+\begin{itemize}
+  \item $743$ distinct $6$-edge cuts examined.
+  \item $1486$ $(\text{graph}, \text{cut}, \text{side})$ triples tested.
+  \item $11{,}477$ cut tires examined.
+  \item \textbf{$0$ tree-structure failures} (no cycles in the
+        parent--child relation under the vertex-overlap heuristic).
+\end{itemize}
+
+The data spans:
+\begin{itemize}
+  \item The $6$ Holton-McKay non-Hamiltonian $38$-vertex cubic plane
+        graphs (their duals are $21$-vertex maximal planar graphs of
+        minimal degree $4$ and vertex-connectivity $3$).
+  \item The dodecahedron ($20$-vertex cubic plane graph, dual of the
+        icosahedron, which is a $12$-vertex $5$-regular maximal
+        planar graph with vertex-connectivity $5$).
+\end{itemize}
+
+Although neither family is strictly ``min degree $5$ with vertex
+connectivity $6$'' (which is incompatible with the maximal-planar
+upper bound on average degree of $6 - 12/|V|$), the test covers
+duals of:
+\begin{enumerate}
+  \item Several internally non-trivial maximal planar graphs (HM
+        duals).
+  \item A min-degree-$5$ maximal planar graph (icosahedron).
+\end{enumerate}
+This is broader than the typical chain pigeonhole test bed.
+
+\section*{Empirical demonstration on Holton-McKay \#0 (detailed)}

 \subsection*{$G'_1$ side ($|S| = 28$, depths $0$ to $7$)}