MineChoice iron-pickaxe training checkpoint
This model has not beaten Minecraft. No held-out full-game evaluation or server-verified Ender Dragon kill has been demonstrated.
This is immutable local reward-training checkpoint 59 from Minecraft Java 1.20.4 survival play. The 73-input, 45-action MLP uses direct structured state, not pixels. Coded skills execute categorical decisions, with no teacher fallback at inference. The classifier is not a standalone game bot.
Episode 20260928-030607-train-persistent-001 continued a naturally generated world with server-preserved
inventory. A gameplay transition verified a newly crafted iron pickaxe and its
reward. The episode then stopped at its configured iron-pickaxe milestone, and
its 51 reward transitions trained this checkpoint. This is a training milestone,
not a fresh empty-inventory run, a measured success rate, or proof this
checkpoint will repeat it. See milestone_evidence.json for exact scope.
model.safetensors and policy.json contain the same neural weights.
config.json records input and action order. The bundled policy schema is the
episode's source snapshot. Training ran locally on an RTX 3090; inference does
not require a hosted service.
This checkpoint has not demonstrated diamond acquisition, Nether entry, End entry, or game completion. No world save, credential, or raw gameplay log is included.