{"abstract":"Milestone decay counts milestones after the requested step.","category":"Learning rate schedules","checks":6,"contract":"Multiply base by gamma once for each milestone <= step.","contract_signature":"step, base, gamma, milestones","evaluation_group":"lr-schedule-milestone-decay","failed_approach":"Shifting the comparison forward applies a milestone before its configured step.","family":"w3-learning-rate-schedules-milestone-decay","id":"FA-99066","implementations":{"attempt":{"sha256":"782c64af694c2725d7be151bc645c1ad5b4d61d7847ff7632abfd0bb9435df87","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(step, base, gamma, milestones):\n    return base*gamma**sum(m<=step+1 for m in milestones)\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nms=[2,4,7]; check('at milestone',solve(2,16,0.5,ms),8); check('before',solve(1,16,0.5,ms),16); check('between',solve(3,16,0.5,ms),8); check('second',solve(4,16,0.5,ms),4); check('empty',solve(9,7,0.2,[]),7); check('past all',solve(8,16,0.5,ms),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":"04ec1a7e3ffff3f78a304895b9a93b7973fcaa5f257e6a856630ca70f7bf2cb2","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(step, base, gamma, milestones):\n    return base*gamma**sum(m<step for m in milestones)\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nms=[2,4,7]; check('at milestone',solve(2,16,0.5,ms),8); check('before',solve(1,16,0.5,ms),16); check('between',solve(3,16,0.5,ms),8); check('second',solve(4,16,0.5,ms),4); check('empty',solve(9,7,0.2,[]),7); check('past all',solve(8,16,0.5,ms),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":"These are toy scalar contracts with integer steps and finite parameters; they do not model framework defaults or optimizer state. 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-learning-rate-schedules-milestone-decay","generated_at":"2026-09-30T16:48:19.909905+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A deterministic scalar schedule makes step-to-rate mapping explicit for a small training loop.","root_cause":"Milestone comparisons exclude the exact update index.","sha256":"39c250784776556f52b24f489ad1788f9afcb99efaa9a80423c4a8d9d40563b2","title":"Milestone decay counts milestones after the requested step · case 01","variant":1,"variant_policy":"The same explicitly authored fixtures repeat across all five numbered records; runtime N is unused. These records share one mechanism and must remain together in evaluation splits.","verified":true,"visibility":"public","verification":{"attempt":{"elapsed_ms":28.666,"exit_code":1,"observations":[{"actual":8.0,"check":"at milestone","expected":8,"passed":true},{"actual":8.0,"check":"before","expected":16,"passed":false},{"actual":4.0,"check":"between","expected":8,"passed":false},{"actual":4.0,"check":"second","expected":4,"passed":true},{"actual":7.0,"check":"empty","expected":7,"passed":true},{"actual":2.0,"check":"past all","expected":2,"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"at milestone\", \"actual\": 8.0, \"expected\": 8, \"passed\": true}, {\"check\": \"before\", \"actual\": 8.0, \"expected\": 16, \"passed\": false}, {\"check\": \"between\", \"actual\": 4.0, \"expected\": 8, \"passed\": false}, {\"check\": \"second\", \"actual\": 4.0, \"expected\": 4, \"passed\": true}, {\"check\": \"empty\", \"actual\": 7.0, \"expected\": 7, \"passed\": true}, {\"check\": \"past all\", \"actual\": 2.0, \"expected\": 2, \"passed\": true}], \"passed\": false}\n"},"broken":{"elapsed_ms":28.518,"exit_code":1,"observations":[{"actual":16.0,"check":"at milestone","expected":8,"passed":false},{"actual":16.0,"check":"before","expected":16,"passed":true},{"actual":8.0,"check":"between","expected":8,"passed":true},{"actual":8.0,"check":"second","expected":4,"passed":false},{"actual":7.0,"check":"empty","expected":7,"passed":true},{"actual":2.0,"check":"past all","expected":2,"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"at milestone\", \"actual\": 16.0, \"expected\": 8, \"passed\": false}, {\"check\": \"before\", \"actual\": 16.0, \"expected\": 16, \"passed\": true}, {\"check\": \"between\", \"actual\": 8.0, \"expected\": 8, \"passed\": true}, {\"check\": \"second\", \"actual\": 8.0, \"expected\": 4, \"passed\": false}, {\"check\": \"empty\", \"actual\": 7.0, \"expected\": 7, \"passed\": true}, {\"check\": \"past all\", \"actual\": 2.0, \"expected\": 2, \"passed\": true}], \"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."}}