{ "credit": "Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved: original replica, Gaussian filter, fee rule, qm tables and exact-arithmetic arb/retail tell.", "mech": "age-fee: symmetric saturating additive fee premium since last callback inferred as arbitrage", "already_present": { "value": false, "where": "Inspected leanbase.py Policy.on_grid, tellpol.py TellPolicy.on_trade, real.py RealPolicy.on_trade/lookup and stalerival.py Policy.on_trade from the packed reference. real.py grows filter variance over elapsed steps; stalerival.py AGE varies a rival-price pull using time since a router-binding retail callback. Neither keeps a last-inferred-arbitrage clock or adds a symmetric age fee. No substitute mechanism was necessary.", "source": "https://api.imd.fun/artifacts/efb8e653f8fd061cecf3e92b186470c9c96c5188523364b60ed101e9672869aa" }, "definition": { "formula": "On each callback: if existing is_arb (step-first and posterior > 0.5), last_arb_t = t; age = max(0,t-last_arb_t); d = premium_bps * 1e-4 * age / (age + age_tau); bid = fee_round(base_bid+d); ask = fee_round(base_ask+d).", "parameters": { "premium_bps": "Nonnegative asymptotic add-on in basis points, grid [0.1,0.3,1.0,3.0].", "age_tau": "Positive half-saturation time in simulator timestamp units, grid [1,4,16]." }, "state_kept": "One last_arb_t timestamp, initialized to 0 at the simulation origin. No external arb flag, fair price, rival reserves, run volatility or arrival rate is used. Time before the first inferred arb counts from origin. Same-timestamp retail callbacks after an inferred arb have zero premium. Quotes change only on callbacks, not during silence.", "integration": "Added after the existing filter, tell, rival pull and table lookup, before fee rounding/clamp; both corrected fees are persisted as fees in force for subsequent inference. No neural hook weights. AGE retains independent router-age state and uses a0=0, ainf=0.7, tau=1. LEAN uses a0=ainf=0.35, tau=2 (irrelevant when a0=ainf).", "experiment": "12 predeclared Cartesian settings; 10000 steps per seed. Selection 64000..65023 in consecutive 128-seed rounds, paired with original LEAN. Drop immediately when cumulative mean_diff+3*se<0; choose largest mean among full-budget survivors (if none, largest measured mean and disclose). Choice frozen before reports. Each report uses 10000..12047 once per arm; no report retuning.", "standard_error": "sample standard deviation of per-seed candidate-minus-baseline edge differences, ddof=1, divided by sqrt(n). seeds_positive counts strictly positive differences.", "reproduction": "Apply patch to fastlean/fastlean.py from source artifact; install numpy==2.2.6, numba==0.61.2, llvmlite==0.44.0 with Python 3.12. Load the pinned pol_mid.json and qm_k1e-5.json. fastlean.run_batch(fastlean.load(amm_dir,premium_bps=c,age_tau=tau,**baseline_options),range(seed0,seed0+n)) produces candidate edges; omit premium parameters for the baseline.", "sources": [ "https://api.imd.fun/artifacts/a88ff982eba7e078bc6ccb1555b90a211c00c9fd76ec90a62ab13a4074f2c3d0", "https://api.imd.fun/artifacts/efb8e653f8fd061cecf3e92b186470c9c96c5188523364b60ed101e9672869aa", "https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/solved/SHA256SUMS" ], "sha256": { "original_fastlean.py": "d9c758bb280fe24e09c039496f84ab0804173ff70cec24c332f8683f28efa582", "patched_fastlean.py": "2e19cd4c260143d499249a8a325beee0e9293b73ec9c348f2a4a7f7d600335b4", "unified_diff": "9705f4f89ce433af01d552c5081b176a95fc2639f53b573b25aa0d3fc2f2bb3c", "pol_mid.json": "efbb5ee102db4d312f64565d9bf7232b6becbe02b624164f744a402bbbf2983b", "qm_k1e-5.json": "5ff53e42ca87390e5f65d1caa3f7a480e67a8212fb0670274dfd562b05ba1afd", "selection_record": "947505ad7646c2b576a64beaa9ce43777d2661d5bc0ce24c299b77f6a1460e0d", "report_record": "ac02e6a36178cef0be2bd053b951a27b81e06475bbbd23a0ce4497c25f44c57c" }, "checks": { "zero_premium": { "credit": "Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved: replica, Gaussian filter, fee rule, qm tables, arb/retail tell", "zero_premium_parity_n": 128, "max_abs_diff": 0.0 }, "independent_cpython_reference": { "credit": "Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved: replica, Gaussian filter, fee rule, qm tables, arb/retail tell", "checks": [ { "parameters": { "premium_bps": 0.1, "age_tau": 16.0 }, "on_age": false, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 }, { "parameters": { "premium_bps": 0.1, "age_tau": 16.0 }, "on_age": true, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 }, { "parameters": { "premium_bps": 3.0, "age_tau": 1.0 }, "on_age": false, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 }, { "parameters": { "premium_bps": 3.0, "age_tau": 1.0 }, "on_age": true, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 } ] }, "chosen_cpython_reference": { "credit": "Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved: replica, Gaussian filter, fee rule, qm tables, arb/retail tell", "checks": [ { "parameters": { "premium_bps": 0.1, "age_tau": 4.0 }, "on_age": false, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 }, { "parameters": { "premium_bps": 0.1, "age_tau": 4.0 }, "on_age": true, "n": 4, "max_abs_diff": 0.0, "bit_identical": 4 } ] }, "reference_seeds": [ 64000, 64001, 64002, 64003 ], "reference_module_hashes": "All 10 packed modules listed in frozen.dependency_sha256 match. experiment.py is packed but has no hash in that mapping and is not imported for scoring. Both table hashes match the pinned SHA256SUMS.", "scope": "Local replica implementation checks; not engine or Solidity gate certification." } }, "settings_tried": [ { "parameters": { "premium_bps": 0.1, "age_tau": 1.0 }, "dropped_early": false, "rounds": [ { "n": 128, "mean_diff": 0.012352516473067787, "se": 0.011269405787603786, "seeds_positive": 69 }, { "n": 256, "mean_diff": 0.008704156566248233, "se": 0.006991258823388255, "seeds_positive": 135 }, { "n": 384, "mean_diff": 0.0026189650433581235, "se": 0.005677459509799799, "seeds_positive": 188 }, { "n": 512, "mean_diff": 0.0019552730148637565, "se": 0.004890391814250944, "seeds_positive": 251 }, { "n": 640, "mean_diff": -0.002470079548591375, "se": 0.004441573855229731, "seeds_positive": 310 }, { "n": 768, "mean_diff": -0.000788844342967554, "se": 0.004029940458611967, "seeds_positive": 376 }, { "n": 896, "mean_diff": -0.0018342010734418196, "se": 0.0037925992073618462, "seeds_positive": 437 }, { "n": 1024, "mean_diff": -0.0022532207388445413, "se": 0.0034551722989029413, "seeds_positive": 495 } ], "n": 1024, "mean_diff": -0.0022532207388445413, "se": 0.0034551722989029413, "seeds_positive": 495 }, { "parameters": { "premium_bps": 0.1, "age_tau": 4.0 }, "dropped_early": false, "rounds": [ { "n": 128, "mean_diff": -0.0012489227085441534, "se": 0.00917456584516074, "seeds_positive": 54 }, { "n": 256, "mean_diff": -0.002036558802306132, "se": 0.006076665023713805, "seeds_positive": 124 }, { "n": 384, "mean_diff": -2.063288902827883e-05, "se": 0.005112556371248917, "seeds_positive": 190 }, { "n": 512, "mean_diff": -0.0001826283941797069, "se": 0.004541459414191794, "seeds_positive": 251 }, { "n": 640, "mean_diff": -0.0024899574770905096, "se": 0.004041818259447052, "seeds_positive": 309 }, { "n": 768, "mean_diff": -0.0011334108429589633, "se": 0.0036012287672035747, "seeds_positive": 368 }, { "n": 896, "mean_diff": -0.0016263393136807405, "se": 0.0033700797132348863, "seeds_positive": 432 }, { "n": 1024, "mean_diff": -0.00098988062173061, "se": 0.0030843479989459822, "seeds_positive": 492 } ], "n": 1024, "mean_diff": -0.00098988062173061, "se": 0.0030843479989459822, "seeds_positive": 492 }, { "parameters": { "premium_bps": 0.1, "age_tau": 16.0 }, "dropped_early": false, "rounds": [ { "n": 128, "mean_diff": 0.003576141338796912, "se": 0.008028463240833077, "seeds_positive": 64 }, { "n": 256, "mean_diff": -0.004695243088218781, "se": 0.0061279544114613066, "seeds_positive": 124 }, { "n": 384, "mean_diff": -0.004388536692842522, "se": 0.005018184302192923, "seeds_positive": 188 }, { "n": 512, "mean_diff": -0.002447375987575051, "se": 0.00414398388580654, "seeds_positive": 265 }, { "n": 640, "mean_diff": -0.0035115533130853473, "se": 0.0037373771791540237, "seeds_positive": 330 }, { "n": 768, "mean_diff": -0.0010847182685730743, "se": 0.0032689179677714607, "seeds_positive": 401 }, { "n": 896, "mean_diff": -0.0015265148993693117, "se": 0.003107249061412016, "seeds_positive": 470 }, { "n": 1024, "mean_diff": -0.0012569996153170004, "se": 0.0028768428370276964, "seeds_positive": 533 } ], "n": 1024, "mean_diff": -0.0012569996153170004, "se": 0.0028768428370276964, "seeds_positive": 533 }, { "parameters": { "premium_bps": 0.3, "age_tau": 1.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.014210128102622566, "se": 0.01374503345460881, "seeds_positive": 60 }, { "n": 256, "mean_diff": -0.013599619407643537, "se": 0.009513608316024342, "seeds_positive": 117 }, { "n": 384, "mean_diff": -0.013729903762747261, "se": 0.007585043764637116, "seeds_positive": 177 }, { "n": 512, "mean_diff": -0.017116274706807433, "se": 0.006446634291439931, "seeds_positive": 241 }, { "n": 640, "mean_diff": -0.021789774573505928, "se": 0.005868017206595624, "seeds_positive": 294 } ], "n": 640, "mean_diff": -0.021789774573505928, "se": 0.005868017206595624, "seeds_positive": 294 }, { "parameters": { "premium_bps": 0.3, "age_tau": 4.0 }, "dropped_early": false, "rounds": [ { "n": 128, "mean_diff": -0.004682068359308289, "se": 0.012047334186646979, "seeds_positive": 65 }, { "n": 256, "mean_diff": -0.0018474852393290941, "se": 0.008103950995154534, "seeds_positive": 126 }, { "n": 384, "mean_diff": -0.005239053982061777, "se": 0.0064559145679925515, "seeds_positive": 185 }, { "n": 512, "mean_diff": -0.0032032578986311844, "se": 0.005615758024719256, "seeds_positive": 250 }, { "n": 640, "mean_diff": -0.007660640852016698, "se": 0.0050684364067833665, "seeds_positive": 310 }, { "n": 768, "mean_diff": -0.004081463988117188, "se": 0.0049174457619981715, "seeds_positive": 376 }, { "n": 896, "mean_diff": -0.004140411864042903, "se": 0.004644375942463013, "seeds_positive": 441 }, { "n": 1024, "mean_diff": -0.0023674191138677303, "se": 0.004249355203770388, "seeds_positive": 506 } ], "n": 1024, "mean_diff": -0.0023674191138677303, "se": 0.004249355203770388, "seeds_positive": 506 }, { "parameters": { "premium_bps": 0.3, "age_tau": 16.0 }, "dropped_early": false, "rounds": [ { "n": 128, "mean_diff": 0.00024285375621602157, "se": 0.008220008532406866, "seeds_positive": 56 }, { "n": 256, "mean_diff": -0.0019177064701612068, "se": 0.005707703036432349, "seeds_positive": 119 }, { "n": 384, "mean_diff": -0.005875735093548329, "se": 0.005224694155268137, "seeds_positive": 171 }, { "n": 512, "mean_diff": -0.005335052077045543, "se": 0.0045198202244325905, "seeds_positive": 235 }, { "n": 640, "mean_diff": -0.005914532173983478, "se": 0.004132656555231623, "seeds_positive": 300 }, { "n": 768, "mean_diff": -0.004486576279836951, "se": 0.0038288479339564294, "seeds_positive": 369 }, { "n": 896, "mean_diff": -0.005203402951248469, "se": 0.003625269980421159, "seeds_positive": 435 }, { "n": 1024, "mean_diff": -0.004005399044143765, "se": 0.0033521141051722865, "seeds_positive": 500 } ], "n": 1024, "mean_diff": -0.004005399044143765, "se": 0.0033521141051722865, "seeds_positive": 500 }, { "parameters": { "premium_bps": 1.0, "age_tau": 1.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.09602398276018276, "se": 0.02218983216289735, "seeds_positive": 46 } ], "n": 128, "mean_diff": -0.09602398276018276, "se": 0.02218983216289735, "seeds_positive": 46 }, { "parameters": { "premium_bps": 1.0, "age_tau": 4.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.044231825353639564, "se": 0.017266370463604616, "seeds_positive": 56 }, { "n": 256, "mean_diff": -0.060467783255727436, "se": 0.012928942300628571, "seeds_positive": 103 } ], "n": 256, "mean_diff": -0.060467783255727436, "se": 0.012928942300628571, "seeds_positive": 103 }, { "parameters": { "premium_bps": 1.0, "age_tau": 16.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.01346266515434591, "se": 0.012082211432524439, "seeds_positive": 59 }, { "n": 256, "mean_diff": -0.009304998302386513, "se": 0.008503958379930622, "seeds_positive": 124 }, { "n": 384, "mean_diff": -0.012757268326150909, "se": 0.0071099070333835454, "seeds_positive": 177 }, { "n": 512, "mean_diff": -0.011190140719975594, "se": 0.006078411179754858, "seeds_positive": 243 }, { "n": 640, "mean_diff": -0.016742500477311406, "se": 0.005408988701739587, "seeds_positive": 294 } ], "n": 640, "mean_diff": -0.016742500477311406, "se": 0.005408988701739587, "seeds_positive": 294 }, { "parameters": { "premium_bps": 3.0, "age_tau": 1.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.7013991299340345, "se": 0.043207843797234406, "seeds_positive": 8 } ], "n": 128, "mean_diff": -0.7013991299340345, "se": 0.043207843797234406, "seeds_positive": 8 }, { "parameters": { "premium_bps": 3.0, "age_tau": 4.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.3281102803617273, "se": 0.03328474063286448, "seeds_positive": 22 } ], "n": 128, "mean_diff": -0.3281102803617273, "se": 0.03328474063286448, "seeds_positive": 22 }, { "parameters": { "premium_bps": 3.0, "age_tau": 16.0 }, "dropped_early": true, "rounds": [ { "n": 128, "mean_diff": -0.09154545579674478, "se": 0.0213959238634437, "seeds_positive": 41 } ], "n": 128, "mean_diff": -0.09154545579674478, "se": 0.0213959238634437, "seeds_positive": 41 } ], "chosen": { "index_zero_based": 1, "parameters": { "premium_bps": 0.1, "age_tau": 4.0 }, "selection_rule": "Highest cumulative paired mean among settings surviving all 1024 selection seeds. If all dropped, highest measured paired mean.", "all_settings_dropped": false }, "report": { "vs_lean": { "n": 2048, "mean_diff": 0.0005244462906790204, "se": 0.0023991125163267874, "seeds_positive": 1062, "baseline_mean": 536.603390956178, "candidate_mean": 536.6039154024688 }, "on_age": { "n": 2048, "mean_diff": -0.001480369005663923, "se": 0.002419455415421662, "seeds_positive": 1051, "baseline_mean": 536.6231002985178, "candidate_mean": 536.621619929512 } }, "promote": false, "contract_cost": { "storage_words": 1, "gas": { "measured": null, "engineering_estimate_added_per_callback": [ 2500, 24000 ], "basis": "Dedicated persistent timestamp: one read on retail callbacks and one timestamp write on inferred arb, plus subtraction, multiply/divide and two additions. Largest estimate allows a cold first zero-to-nonzero timestamp write. Later nonzero writes should be substantially cheaper. Storage packing may change cost. No Solidity implementation compiled or gas measured." }, "characters": { "measured": null, "engineering_estimate_added": [ 400, 1000 ] }, "runtime_bytes": { "measured": null, "engineering_estimate_added": [ 150, 500 ] }, "budget_context": "Requester-supplied current peak 208000/245000 gas, runtime 22440/24576 bytes and source 47230/50000 characters leave 37000 gas, 2136 bytes and 2770 characters. This small arithmetic extension plausibly fits, but estimates are not a gas/size pass or an OOG guarantee." }, "patch_zb64": "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", "not_done": [ "No QLX5 or qt1 comparison: this bounded item evaluates LEAN and AGE on the specified fast replica.", "No engine-faithfulness repair: the supplied replica retains constant K. Engine and leaderboard transfer are unverified.", "No Solidity port, deployed/source size measurement, peak gas check, OOG check, or P/Q/T/F/G/E contract gate certification. Local parity checks are described separately.", "No fresh engine seeds consumed, no pipeline promotion unless both report gains meet the specified threshold, and no leaderboard submission.", "No multiplier, delayed-onset or other mechanisms tested beyond the 12 stated additive-premium settings.", "Per-seed raw edges and installed dependencies are scratch-only research materials, not delivered runtime dependencies. Output is a result record and a compressed source patch, not an executable package." ] }