Each piece switched off in turn, trained at identical compute, then played against the full stack over 8192 duplicate matches. Below 0.5 means the removed piece was doing work. - both seats: 0.3317 [0.322, 0.342]. The biggest single contributor. Half of it is simply sample count -- dropping the opponent seat halves the learner actions per update -- but that is the point: self-play already produced those plies with the same network, and the old trainer stop_gradiented them away. - match observation: 0.4751 [0.464, 0.486]. Small but real. Since carry itself contributes almost nothing (rounds decompose), most of this is likely the single-round observation defects being fixed: to_move, the deck clock, and the score_diff scale. - privileged critic: 0.5160 [0.505, 0.527] -- turning it OFF makes the agent significantly STRONGER. Fable called this the biggest missing idea; it is wrong. A critic that knows the deck fits V(full state), which is not E[return | masked obs], so the advantage picks up a component the actor cannot act on. From the actor's side that is noise, not variance reduction. Asymmetric critics hurting under partial observability is a known failure mode. Defaulted off accordingly. (Reusing it as a PIMC leaf evaluator may still stand -- that is a separate claim from using it to train the policy.) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XBQKgvBbxbheiTF1AVy1Sh
coolrl-lost-cities
JAX PPO training and on-device browser play for the two-player card game Lost Cities. The project provides a pure JAX rules engine, PPO training and evaluation tools, and a static web client that runs the shipped final policy with WebGPU when available and WebAssembly otherwise.
The active path is JAX + PPO. Deep CFR and ISMCTS are retained only as historical research implementations; see legacy notes.
Quick start
Install the project, then run a complete CPU-sized sanity check:
uv sync
uv run lost-cities-jax-ppo rollout-smoke --config configs/jax_ppo/smoke.yaml
uv run lost-cities-jax-ppo train --config configs/jax_ppo/smoke.yaml
The final command prints a run directory containing latest, config.json,
and metrics.jsonl under runs/tmp/jax-ppo-artifacts/.
Train and evaluate a PPO policy
The committed configurations describe the opponent and training budget. For a GPU run, use CUDA JAX and keep generated artifacts outside git:
flock -n .compute.lock uv run --with 'jax[cuda12]' lost-cities-jax-ppo train \
--config configs/jax_ppo/balanced.yaml \
--set run.artifact_root=runs/jax-ppo
Evaluate a saved checkpoint against a fixed opponent. Duplicate evaluation swaps seats over the same shuffled games:
uv run lost-cities-jax-ppo eval \
--config configs/jax_ppo/balanced.yaml \
--checkpoint runs/jax-ppo/<run>/latest \
--opponent heuristic_balanced \
--games 10000 \
--duplicate
Run uv run lost-cities-jax-ppo --help for training against saved opponents,
league runs, gates, human-play logs, and evaluation variants.
Browser client
The final verified JAX PPO policy is shipped as a 3.1 MB static ONNX asset. No server or local checkpoint is required to play it:
cd web
npm ci
npm run dev
npm run build produces a fully static site. The model path is deployment-base
aware, so the build can be served from GitHub Pages, GitLab Pages, or a normal
web root. Pushes to main deploy that build to both configured Pages hosts.
See web/README.md for model replacement, tests, and the
cross-runtime parity fixture.
JAX Rules Engine
lost_cities_jax is a standalone pure rules simulator for one two-player
Lost Cities round. It does not contain neural networks, PPO, CFR, MCTS, match
wrappers, bots, or rule variants.
Public API:
from lost_cities_jax import (
OBS_DIM,
N_ACTIONS,
State,
batched_legal_mask,
batched_obs,
batched_reset,
batched_step,
board_score,
legal_action_mask,
observation,
reset,
reset_from_order,
score,
step,
)
Core functions are pure JAX functions:
reset(rng) -> Statereset_from_order(deck_order) -> Statelegal_action_mask(state) -> bool[96]step(state, action) -> (State, float32[2], bool)score(state) -> float32[2]board_score(state) -> float32[2]observation(state, player) -> float32[454]
The batched exports are jax.jit(jax.vmap(...)) wrappers. Illegal actions and
done-state actions are defined as no-op transitions with zero reward; training
code should still sample only from legal_action_mask.
Rule Summary
One round uses 60 cards: five colors, each with three handshakes and ranks 2 through 10. Each player starts with eight cards, then each ply must place one hand card to the matching expedition or discard pile and draw one card from the deck or a discard pile. A player may not draw the card they just discarded.
Expedition numbers must be strictly increasing. Handshakes may be played only
before any number in that color. The round ends immediately when the final deck
card is drawn, or at MAX_STEPS == 400; forced termination is scored exactly
like natural termination.
Scoring per player/color:
empty column: 0
non-empty: (sum(number ranks) - 20) * (1 + handshake_count)
bonus: +20 if total column length >= 8, not multiplied
Encodings
Cards:
| Field | Encoding |
|---|---|
card_id |
color * 12 + slot |
color |
0..4 |
slot 0..2 |
handshake |
slot 3..11 |
ranks 2..10, with rank = slot - 1 |
Actions (N_ACTIONS == 96):
action_id = hand_slot * 12 + place_type * 6 + draw_source
hand_slot = 0..7, current player's hand sorted by card_id
place_type = 0 play, 1 discard
draw_source = 0 deck, 1..5 discard pile color 0..4
Observation (OBS_DIM == 454):
- 60 cards x 7 one-hot channels: my hand, my board, opponent board, discard top, discard non-top, opponent public hand, unknown.
- 34 scalar features:
remaining deck
/44, opponent unknown hand count/8, step count/400, current-player then opponentcol_top /10,col_hs /3,col_len /12, and current board score difference(player - opponent) /780.
Verification
uv run pytest -q tests/lost_cities_jax
uv run pytest -q
uv run ruff check .
Large differential profiles:
# CI profile: 100,000 random legal-policy games
CI=1 uv run pytest -q tests/lost_cities_jax/test_differential.py
# Full profile: 1,000,000 random legal-policy games
uv run pytest -q tests/lost_cities_jax/test_differential.py --full
Observed differential results on 2026-07-04 with CPU JAX backend:
CI=1 ... test_differential.py
1 passed in 247.61s (0:04:07)
... test_differential.py --full
1 passed in 2514.14s (0:41:54)
elapsed=41:54.48
Throughput benchmark:
flock -n .compute.lock uv run python benchmarks/throughput.py
# Optional CUDA check without making CUDA a project dependency:
flock -n .compute.lock uv run --with 'jax[cuda12]' python benchmarks/throughput.py
Measured on 2026-07-04 with CPU JAX backend:
backend=cpu
batch_size=8192
steps=256
elapsed_sec=4.655526
steps_per_sec=450465.14
Measured on 2026-07-04 with CUDA JAX backend on RTX 3090, using the optional
uv run --with 'jax[cuda12]' ... command:
backend=gpu
batch_size=8192
steps=256
elapsed_sec=0.530039
steps_per_sec=3956598.38
DECISIONS.md
- Explicit
deck_orderdealing uses the first eight cards for player 0 and the next eight for player 1. The remaining cards are drawn from index 16. This is equivalent under a uniform shuffle and is fixed by tests. - After a legal terminal transition,
to_moveis advanced to the next player, butdone=Truemakes all later steps complete no-ops. - Terminal reward is emitted only on the transition that reaches
done=True. Done-state no-op steps return zero reward. - Observation scalar normalization is implementation-defined as documented
above and locked by the exported
OBS_DIM.
JAX PPO Static-Opponent Ladder
The first training stack above lost_cities_jax is exposed as:
uv run lost-cities-jax-ppo --help
CPU smoke:
uv run lost-cities-jax-ppo rollout-smoke --config configs/jax_ppo/smoke.yaml
uv run lost-cities-jax-ppo train --config configs/jax_ppo/smoke.yaml
GPU training uses optional CUDA JAX, keeping CUDA wheels out of the default project dependency set:
tmux new-session -s coolrl-jax-ppo-discard \
-c /home/coolguy/dev/coolrl-lost-cities \
"flock -n .compute.lock uv run --with 'jax[cuda12]' lost-cities-jax-ppo train \
--config configs/jax_ppo/discard-only.yaml"
Random-policy baseline vs discard_only, measured on 2026-07-04 with
8192 games x 400 plies on GPU:
return_mean=-55.582763671875
game_length_mean=69.6085205078125
max_steps_rate=0.0
play_action_rate=0.28854578733444214
opened_colors_mean=4.942626953125
positive_expeditions_mean=0.41796875
Static-opponent gate results, measured on 2026-07-04 with 10,000 fixed shuffles and duplicate seat-swapped evaluation:
| Gate | Opponent | Result | Win rate (Wilson 95%) | Mean score diff | Mean length | Positive expeditions/game |
|---|---|---|---|---|---|---|
| 1 | discard_only |
PASS | 1.00000 [0.99981, 1.00000] | 204.56335 | 82.45495 | 3.2783 |
| 2 | heuristic_balanced |
PASS | 0.98655 [0.98486, 0.98806] | 116.83400 | 167.81450 | 3.9573 |
| 3 | heuristic_cautious |
PASS | 0.95955 [0.95673, 0.96219] | 142.89930 | 185.38690 | 4.0946 |
Large PPO artifacts are written under
/mnt/2tbhdd/coolrl-lost-cities-artifacts/jax-ppo-static-opponents/.
Generated checkpoints and evaluation JSON are not committed to git. The full
run summary is in
docs/reports/jax-ppo-static-opponent-ladder-2026-07-04.md.
Basic Usage
from coolrl_lost_cities.games.classic import GameState, build_bot, classic_config
state = GameState.new_game(classic_config(seed=1))
bot = build_bot("random", seed=1)
while not state.terminal:
state.apply_action(bot.act(state))
print(state.total_score(0), state.total_score(1))
See classic port notes for the current direction.