Notebook #078 · published

DeFi maths, Haiku 5.5 with thinking off

Exact means the final number is right to the cent (to the whole token for bonding-curve buys). Each part ran against the mean of two unchanged runs of the same agent, on 432 paired tasks drawn by a public beacon after the design was hashed. The unchanged agent was exact on 59.4% of them. Rule v2: every interval is widened for the 7 arms this notebook compares, so luck flags any of them at most 5% of the time.

Calculator tool + Double-check, Calculator tool + A penalty threat, Calculator tool + Read the question again, Verify by inverse and Decompose and verify products helped. 1 part is not proven either way yet. OPSIN called 5 of 6 before the beacon drew the tasks. Its foresight, committed before the beacon, gave what came back 65% on average; a guess gives 25%.

Preregistered
sha256 7ffe05a2955b67e72751a271e03a41de334d3ba070d5dbcddc5765082ad27fb0
the full design, as JSON
Tasks drawn by
drand round 6,624,158
seed drand:6624158:4ec8f6a44cdd87d2f828a946d52c7f82e185c594a656f44c819b1f61d1793583
Signed by
0xFf8E1E8ebF21064Ef781B75CdC8DFbd969399EbD
EIP-191 over sha256 of each verdict’s canonical JSON. Verify one yourself
Every run
Each run of every arm, with its task’s family and whether it was exact: CSV or JSON. With the seed, every verdict above can be recomputed from them.
Recorded on Elysium
transaction 0x83882b5e58a90568348d3c2ced5274230c4edd8b6e8d56ba84d15c5e74f45f63
Written once, by the lab, and nothing can change it. The contract
  1. Control check (A/A)the unchanged agent against itselfsigned verdict 483
    UNRESOLVED +3.9−0.5 to +8.3
  2. PlaceboThe weather outside is mild today.signed verdict 484
    UNRESOLVED −2.2−8.0 to +3.4 ×0.94
  3. Calculator tool + Double-checkstack · OPSIN's replication · OPSIN predicted EXCITES at 0.93 · foresight EXCITES 100% · stopped early at 48 · provisional: stopped early, confirmed only at full size · against the placebo +37.5signed verdict 485
    EXCITES +41.7+25.0 to +58.3 ×1.85
  4. Calculator tool + A penalty threatstack · OPSIN's replication · OPSIN predicted EXCITES at 0.93 · foresight EXCITES 100% · stopped early at 48 · provisional: stopped early, confirmed only at full size · against the placebo +37.5signed verdict 486
    EXCITES +41.7+25.0 to +58.3 ×1.46
  5. Calculator tool + Read the question againstack · OPSIN's replication · OPSIN predicted EXCITES at 0.93 · foresight EXCITES 100% · stopped early at 48 · provisional: stopped early, confirmed only at full size · against the placebo +37.5signed verdict 487
    EXCITES +41.7+25.0 to +58.3 ×1.59
  6. Full-precision intermediatesline · OPSIN's design, from the misses it read · OPSIN predicted UNRESOLVED at 0.40 · foresight UNRESOLVED 54%, EXCITES 19%, INHIBITS 19% · against the placebo +5.6signed verdict 488
    UNRESOLVED +3.4−2.4 to +9.3 ×1.21
  7. Verify by inverseline · OPSIN's design, from the misses it read · OPSIN predicted UNRESOLVED at 0.40 · foresight UNRESOLVED 54%, EXCITES 20%, INHIBITS 19% · stopped early at 48 · provisional: stopped early, confirmed only at full size · against the placebo +20.8signed verdict 489
    EXCITES +25.0+10.4 to +40.6 ×2.03
  8. Decompose and verify productsline · OPSIN's follow-up · OPSIN predicted EXCITES at 0.60 · foresight EXCITES 68%, UNRESOLVED 21% · stopped early at 48 · provisional: stopped early, confirmed only at full size · against the placebo +18.8signed verdict 490
    EXCITES +22.9+6.3 to +40.6 ×2.08