id: 860a657faa2a479ea84be711c666573e
parent_id: dfa96c24f57d4c14bfdb546e402398ad
item_type: 1
item_id: f3c42f4435644e19b186265dd3fbf672
item_updated_time: 1787161239920
title_diff: "[]"
body_diff: "[{\"diffs\":[[0,\" #8 \"],[-1,\"PROPOSED\"],[1,\"= **GEN 6**\"],[0,\": Op\"]],\"start1\":226,\"start2\":226,\"length1\":16,\"length2\":19},{\"diffs\":[[0,\"us: \"],[-1,\"IDEA — candidate for Gen 7 / post-v19. Proposed by Jan\"],[1,\"VISION for next generation. Proposed by Jan. Renamed 2026-08-19: old \\\"Gen 6\\\" (PreflopNet/PPO, −644 BB/100, abandoned) is dropped from the numbering — this is now Gen 6\"],[0,\".**\\\n\"]],\"start1\":305,\"start2\":305,\"length1\":62,\"length2\":175},{\"diffs\":[[0,\"bels).\\\n\\\n\"],[1,\"## NN for speed (Jan, 2026-08-19)\\\nNN inference is ~µs — the natural tool to replace the expensive part (rollout+MC)\\\nwith a fast learned evaluator. Two slots:\\\n- **Distilled EV/action net** = live plays a fast net trained on offline rollout\\\n  labels (revived EquityNet pattern). Primary plan.\\\n- Later: rollout-policy net (cheap action sampler inside the rollout itself,\\\n  MCTS-policy-style) if full oracle sampling inside rollouts is too slow even offline.\\\n\\\n\"],[0,\"## Sugge\"]],\"start1\":1929,\"start2\":1929,\"length1\":16,\"length2\":472},{\"diffs\":[[0,\"staging\\\n\"],[-1,\"\"],[0,\"1. Oracl\"]],\"start1\":2406,\"start2\":2406,\"length1\":16,\"length2\":16},{\"diffs\":[[0,\"m side)\\\n\"],[1,\"\"],[0,\"3. A/B a\"]],\"start1\":2540,\"start2\":2540,\"length1\":16,\"length2\":16},{\"diffs\":[[0,\" as \"],[-1,\"g56-style \"],[0,\"cand\"]],\"start1\":2554,\"start2\":2554,\"length1\":18,\"length2\":8}]"
metadata_diff: {"new":{},"deleted":[]}
encryption_cipher_text: 
encryption_applied: 0
updated_time: 2026-08-19T17:46:05.724Z
created_time: 2026-08-19T17:46:05.724Z
is_locked: 0
type_: 13