{"abstract":"The reduction disagrees with its explicit aggregation oracle.","category":"Numerical aggregation","checks":9,"contract":"Rows [group label, integer residual or None] are reduced into mean squared residual per group, then arithmetic mean across observed groups. Missing residuals contribute neither count nor group presence; group labels are case-sensitive. Empty observed population returns None. Exact Fraction string.","evaluation_group":"s3-na-group-balanced-squared-loss","failed_approach":"Choosing the smallest group also excludes the other equally weighted groups.","family":"s3-numerical-aggregation-group-balanced-squared-loss-macro-largest-group","id":"FA-14221","implementations":{"attempt":{"sha256":"a2ff85c0ed03ed7d57dfe628508a14e8d10a5c275151057fd4870e260b20bbbc","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\nfrom fractions import Fraction\nfrom collections import Counter, defaultdict\nimport math\nimport itertools\nN = 1\nobservations = []\ndef solve(rows):\n    groups={}\n    for label,error in rows:\n        if error is None: continue\n        groups.setdefault(label,[]).append(error)\n    if not groups: return None\n    values=min(groups.values(),key=len)\n    means=[Fraction(sum(e*e for e in values),len(values))]\n    return str(sum(means)/len(means))\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('regression 1', solve(*([('a', 1), ('a', 2), ('b', 4)],)), '37/4')\ncheck('regression 2', solve(*([('a', 1), ('a', 3), ('b', 4)],)), '21/2')\ncheck('regression 3', solve(*([],)), None)\ncheck('regression 4', solve(*([('a', None), ('b', None)],)), None)\ncheck('regression 5', solve(*([('a', 0), ('a', 2), ('b', -3), ('b', 1)],)), '7/2')\ncheck('regression 6', solve(*([('a', 1), ('b', 2), ('b', 2), ('b', 2)],)), '5/2')\ncheck('regression 7', solve(*([('a', None), ('a', 2), ('b', 4)],)), '10')\ncheck('regression 8', solve(*([('A', 1), ('a', 4), ('B', 2)],)), '7')\ncheck(\"variable group balance\",solve([(\"a\",N),(\"a\",-N),(\"b\",2*N)]),str(Fraction(5*N*N,2)))\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"},"broken":{"sha256":"189beee13447f815c1220577f309fd84b54d0555ccbc24f4db89f61c5d094519","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\nfrom fractions import Fraction\nfrom collections import Counter, defaultdict\nimport math\nimport itertools\nN = 1\nobservations = []\ndef solve(rows):\n    groups={}\n    for label,error in rows:\n        if error is None: continue\n        groups.setdefault(label,[]).append(error)\n    if not groups: return None\n    values=max(groups.values(),key=len)\n    means=[Fraction(sum(e*e for e in values),len(values))]\n    return str(sum(means)/len(means))\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('regression 1', solve(*([('a', 1), ('a', 2), ('b', 4)],)), '37/4')\ncheck('regression 2', solve(*([('a', 1), ('a', 3), ('b', 4)],)), '21/2')\ncheck('regression 3', solve(*([],)), None)\ncheck('regression 4', solve(*([('a', None), ('b', None)],)), None)\ncheck('regression 5', solve(*([('a', 0), ('a', 2), ('b', -3), ('b', 1)],)), '7/2')\ncheck('regression 6', solve(*([('a', 1), ('b', 2), ('b', 2), ('b', 2)],)), '5/2')\ncheck('regression 7', solve(*([('a', None), ('a', 2), ('b', 4)],)), '10')\ncheck('regression 8', solve(*([('A', 1), ('a', 4), ('B', 2)],)), '7')\ncheck(\"variable group balance\",solve([(\"a\",N),(\"a\",-N),(\"b\",2*N)]),str(Fraction(5*N*N,2)))\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"},"fixed":{"sha256":"37d21a2cc900138bb57366e9a3e5fcf9a8544a6c56d61385583ab74a4b2e6dd8","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\nfrom fractions import Fraction\nfrom collections import Counter, defaultdict\nimport math\nimport itertools\nN = 1\nobservations = []\ndef solve(rows):\n    groups={}\n    for label,error in rows:\n        if error is None: continue\n        groups.setdefault(label,[]).append(error)\n    if not groups: return None\n    means=[Fraction(sum(e*e for e in values),len(values)) for values in groups.values()]\n    return str(sum(means)/len(means))\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('regression 1', solve(*([('a', 1), ('a', 2), ('b', 4)],)), '37/4')\ncheck('regression 2', solve(*([('a', 1), ('a', 3), ('b', 4)],)), '21/2')\ncheck('regression 3', solve(*([],)), None)\ncheck('regression 4', solve(*([('a', None), ('b', None)],)), None)\ncheck('regression 5', solve(*([('a', 0), ('a', 2), ('b', -3), ('b', 1)],)), '7/2')\ncheck('regression 6', solve(*([('a', 1), ('b', 2), ('b', 2), ('b', 2)],)), '5/2')\ncheck('regression 7', solve(*([('a', None), ('a', 2), ('b', 4)],)), '10')\ncheck('regression 8', solve(*([('A', 1), ('a', 4), ('B', 2)],)), '7')\ncheck(\"variable group balance\",solve([(\"a\",N),(\"a\",-N),(\"b\",2*N)]),str(Fraction(5*N*N,2)))\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"}},"limitations":"Small offline integer/rational inputs only; no performance, statistical inference, or production-library conformance claim. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.","method":"Deterministic executable model with adversarial boundary fixtures.","provenance":{"created_by":"Failure Map","dependencies":"Python standard library","family":"s3-numerical-aggregation-group-balanced-squared-loss-macro-largest-group","generated_at":"2026-09-29T14:39:14.802734+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"Exact bounded examples isolate a reduction defect without floating-point or external-service effects.","repair":"Preserve the group balanced squared loss contract at the identified reduction decision.","root_cause":"Only the largest group is evaluated.","sha256":"d9d5dc4319e7c21a41a59fc11fd16a14706f3536a69ad35347697ae196e4a487","title":"Group balanced squared loss: Only the largest group is evaluated. · case 01","variant":1,"variant_policy":"Five numbered records share a model and may reuse boundary fixtures.","verification":{"attempt":{"elapsed_ms":43.074,"exit_code":1,"observations":[{"actual":"16","check":"regression 1","expected":"37/4","passed":false},{"actual":"16","check":"regression 2","expected":"21/2","passed":false},{"actual":null,"check":"regression 3","expected":null,"passed":true},{"actual":null,"check":"regression 4","expected":null,"passed":true},{"actual":"2","check":"regression 5","expected":"7/2","passed":false},{"actual":"1","check":"regression 6","expected":"5/2","passed":false},{"actual":"4","check":"regression 7","expected":"10","passed":false},{"actual":"1","check":"regression 8","expected":"7","passed":false},{"actual":"4","check":"variable group balance","expected":"5/2","passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"regression 1\", \"actual\": \"16\", \"expected\": \"37/4\", \"passed\": false}, {\"check\": \"regression 2\", \"actual\": \"16\", \"expected\": \"21/2\", \"passed\": false}, {\"check\": \"regression 3\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 4\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 5\", \"actual\": \"2\", \"expected\": \"7/2\", \"passed\": false}, {\"check\": \"regression 6\", \"actual\": \"1\", \"expected\": \"5/2\", \"passed\": false}, {\"check\": \"regression 7\", \"actual\": \"4\", \"expected\": \"10\", \"passed\": false}, {\"check\": \"regression 8\", \"actual\": \"1\", \"expected\": \"7\", \"passed\": false}, {\"check\": \"variable group balance\", \"actual\": \"4\", \"expected\": \"5/2\", \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":46.618,"exit_code":1,"observations":[{"actual":"5/2","check":"regression 1","expected":"37/4","passed":false},{"actual":"5","check":"regression 2","expected":"21/2","passed":false},{"actual":null,"check":"regression 3","expected":null,"passed":true},{"actual":null,"check":"regression 4","expected":null,"passed":true},{"actual":"2","check":"regression 5","expected":"7/2","passed":false},{"actual":"4","check":"regression 6","expected":"5/2","passed":false},{"actual":"4","check":"regression 7","expected":"10","passed":false},{"actual":"1","check":"regression 8","expected":"7","passed":false},{"actual":"1","check":"variable group balance","expected":"5/2","passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"regression 1\", \"actual\": \"5/2\", \"expected\": \"37/4\", \"passed\": false}, {\"check\": \"regression 2\", \"actual\": \"5\", \"expected\": \"21/2\", \"passed\": false}, {\"check\": \"regression 3\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 4\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 5\", \"actual\": \"2\", \"expected\": \"7/2\", \"passed\": false}, {\"check\": \"regression 6\", \"actual\": \"4\", \"expected\": \"5/2\", \"passed\": false}, {\"check\": \"regression 7\", \"actual\": \"4\", \"expected\": \"10\", \"passed\": false}, {\"check\": \"regression 8\", \"actual\": \"1\", \"expected\": \"7\", \"passed\": false}, {\"check\": \"variable group balance\", \"actual\": \"1\", \"expected\": \"5/2\", \"passed\": false}], \"passed\": false}\n"},"fixed":{"elapsed_ms":48.35,"exit_code":0,"observations":[{"actual":"37/4","check":"regression 1","expected":"37/4","passed":true},{"actual":"21/2","check":"regression 2","expected":"21/2","passed":true},{"actual":null,"check":"regression 3","expected":null,"passed":true},{"actual":null,"check":"regression 4","expected":null,"passed":true},{"actual":"7/2","check":"regression 5","expected":"7/2","passed":true},{"actual":"5/2","check":"regression 6","expected":"5/2","passed":true},{"actual":"10","check":"regression 7","expected":"10","passed":true},{"actual":"7","check":"regression 8","expected":"7","passed":true},{"actual":"5/2","check":"variable group balance","expected":"5/2","passed":true}],"passed":true,"stderr":"","stdout":"{\"observations\": [{\"check\": \"regression 1\", \"actual\": \"37/4\", \"expected\": \"37/4\", \"passed\": true}, {\"check\": \"regression 2\", \"actual\": \"21/2\", \"expected\": \"21/2\", \"passed\": true}, {\"check\": \"regression 3\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 4\", \"actual\": null, \"expected\": null, \"passed\": true}, {\"check\": \"regression 5\", \"actual\": \"7/2\", \"expected\": \"7/2\", \"passed\": true}, {\"check\": \"regression 6\", \"actual\": \"5/2\", \"expected\": \"5/2\", \"passed\": true}, {\"check\": \"regression 7\", \"actual\": \"10\", \"expected\": \"10\", \"passed\": true}, {\"check\": \"regression 8\", \"actual\": \"7\", \"expected\": \"7\", \"passed\": true}, {\"check\": \"variable group balance\", \"actual\": \"5/2\", \"expected\": \"5/2\", \"passed\": true}], \"passed\": true}\n"}},"verified":true,"visibility":"public"}