{"abstract":"Disjoint gradient contributions are averaged instead of summed.","category":"Distributed training aggregation","checks":6,"contract":"Given equal-length partial gradient vectors for one parameter block, return their elementwise sum; empty input returns [].","contract_signature":"parts","evaluation_group":"w3-dta-gradient-sum","failed_approach":"Taking only the first contribution loses later shard contributions.","family":"w3-distributed-training-aggregation-gradient-sum","id":"FA-99246","implementations":{"attempt":{"sha256":"a0c5414c70ec0582d3a4d1f5c1c3fe82e3c0bf5aa199aef3f1f866939501f3c9","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(parts):\n    return list(parts[0]) if parts else []\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('two workers',solve([[N,2],[3*N,4]]),[4*N,6])\ncheck('negative cancellation',solve([[N],[-N],[2*N]]),[2*N])\ncheck('empty block',solve([]),[])\ncheck('single worker',solve([[N,0]]),[N,0])\ncheck('three partials',solve([[1],[2],[N]]),[N+3])\ncheck('signed coordinates',solve([[-N,2*N],[N,-N]]),[0,N])\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":"647e6cefd913be653eddd40fe879325c2b5bdd8f0dea094588c2e82471b2bfbc","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(parts):\n    return [sum(v[j] for v in parts)/len(parts) for j in range(len(parts[0]))] if parts else []\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('two workers',solve([[N,2],[3*N,4]]),[4*N,6])\ncheck('negative cancellation',solve([[N],[-N],[2*N]]),[2*N])\ncheck('empty block',solve([]),[])\ncheck('single worker',solve([[N,0]]),[N,0])\ncheck('three partials',solve([[1],[2],[N]]),[N+3])\ncheck('signed coordinates',solve([[-N,2*N],[N,-N]]),[0,N])\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":"Inputs are already materialized Python values from a fixed toy collective; no transport, failure recovery, optimizer state, or real model behavior is represented. 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":"w3-distributed-training-aggregation-gradient-sum","generated_at":"2026-09-30T16:48:22.178335+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A toy distributed-training contract makes replica and shard contribution rules explicit without claiming framework compatibility or performance.","root_cause":"Each worker owns a disjoint parameter slice, so their partial gradients are additive contributions.","sha256":"5d017652ac7898b1a2b0e3ce79f6cde4d2ea720a841c6b9acecdb5188f08ac4e","title":"Disjoint gradient contributions are averaged instead of summed · case 01","variant":1,"variant_policy":"The fixture code references N, supplied as 1 through 5. Fixed boundary checks may repeat. These records share one mechanism and must remain together in evaluation splits.","verified":true,"visibility":"public","verification":{"attempt":{"elapsed_ms":31.031,"exit_code":1,"observations":[{"actual":[1,2],"check":"two workers","expected":[4,6],"passed":false},{"actual":[1],"check":"negative cancellation","expected":[2],"passed":false},{"actual":[],"check":"empty block","expected":[],"passed":true},{"actual":[1,0],"check":"single worker","expected":[1,0],"passed":true},{"actual":[1],"check":"three partials","expected":[4],"passed":false},{"actual":[-1,2],"check":"signed coordinates","expected":[0,1],"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"two workers\", \"actual\": [1, 2], \"expected\": [4, 6], \"passed\": false}, {\"check\": \"negative cancellation\", \"actual\": [1], \"expected\": [2], \"passed\": false}, {\"check\": \"empty block\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"single worker\", \"actual\": [1, 0], \"expected\": [1, 0], \"passed\": true}, {\"check\": \"three partials\", \"actual\": [1], \"expected\": [4], \"passed\": false}, {\"check\": \"signed coordinates\", \"actual\": [-1, 2], \"expected\": [0, 1], \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":29.81,"exit_code":1,"observations":[{"actual":[2.0,3.0],"check":"two workers","expected":[4,6],"passed":false},{"actual":[0.6666666666666666],"check":"negative cancellation","expected":[2],"passed":false},{"actual":[],"check":"empty block","expected":[],"passed":true},{"actual":[1.0,0.0],"check":"single worker","expected":[1,0],"passed":true},{"actual":[1.3333333333333333],"check":"three partials","expected":[4],"passed":false},{"actual":[0.0,0.5],"check":"signed coordinates","expected":[0,1],"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"two workers\", \"actual\": [2.0, 3.0], \"expected\": [4, 6], \"passed\": false}, {\"check\": \"negative cancellation\", \"actual\": [0.6666666666666666], \"expected\": [2], \"passed\": false}, {\"check\": \"empty block\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"single worker\", \"actual\": [1.0, 0.0], \"expected\": [1, 0], \"passed\": true}, {\"check\": \"three partials\", \"actual\": [1.3333333333333333], \"expected\": [4], \"passed\": false}, {\"check\": \"signed coordinates\", \"actual\": [0.0, 0.5], \"expected\": [0, 1], \"passed\": false}], \"passed\": false}\n"}},"member_only":{"stages":["fixed"],"fields":["implementations.fixed","verification.fixed","harness","repair"],"note":"The verified repair, its recorded checks, the repair description, and the scoring harness are available to members."}}