18e16819ca
The rung-9 forest review: what was live, the mechanism of all three observations, the audit rows, and the trap hunt before and after from the release container's own checkpoint. The hunt is brain-driven rather than stubbed, which is legitimate here and was not for sections 13 and 14: this branch adds no macro type, so no population is re-dealt and the two arms differ in the macro code alone. Distinct tiles 216 to 286, windows flagged 70 of 73 to 61, windows under four tiles 21 to 5, BACK starts with no list open 175 to 0, MOVE n starts 36 to 68. Four residuals are named rather than buried: BACK is still 359 starts of 830 over open lists, which is row 34's contract; a NEXT/BACK two-cycle closes the run inside one battle, from a move list whose cursor the seam cannot place on its opening frames; a window spent in a battle is flagged by the tile rule whatever happens in it; and the THROW BALL precondition is inert in this run because the save carries a party of one. Row 41 is the loop behind this one: in the before arm the fly answered the nurse's box YES 474 times on one tile of a Pokemon Center for the last eight brain minutes of the run. The after arm never enters it, so it is neither reproduced nor fixed here. It is the next thing to measure. |
||
|---|---|---|
| .forgejo/workflows | ||
| .github/workflows | ||
| apps/stage | ||
| data/fafb-v783 | ||
| docs | ||
| infra | ||
| packages | ||
| services | ||
| tools | ||
| .gitignore | ||
| CLAUDE.md | ||
| CONTRIBUTING.md | ||
| LICENSE | ||
| LICENSES.md | ||
| Makefile | ||
| NOTICE | ||
| package-lock.json | ||
| package.json | ||
| README.md | ||
| ROM-POLICY.md | ||
flybrain
A simulated fruit-fly brain that plays video games. A connectome-constrained spiking network reads the screen, its population rates become controller inputs, and a scalar reward nudges a bounded set of Kenyon-cell to MBON gains.
The library is @flybrain/brain in packages/brain. It holds the connectome dataset format, the
LIF kernel, the plasticity rule, the population-rate readout and the activity-map geometry. It
holds no game, no emulator and no reward rules: those belong to whatever embeds it.
Workspace layout
| Path | Contents |
|---|---|
packages/brain |
the library (@flybrain/brain) |
data/fafb-v783 |
FlyWire-derived browser artifacts (CC BY-NC 4.0) |
tools/ |
the Python builder that regenerates data/ from official Codex exports |
docs/ |
overview, dataset format, model, plasticity, readout, integration, limitations, verification |
services/flysim |
the Rust service: brain, emulator, snapshot feed, control API, checkpoints |
apps/ |
planned: one directory per game demo |
infra/ |
planned: deployment for the 24/7 stream (see docs/streaming-plan.md) |
Quick start
npm ci
npm test
npm run typecheck
76 tests, about 7 seconds. There is no build step. npx tsx packages/brain/examples/node-random-frames.ts 60
runs the full 139,255-neuron brain with the Game Boy readout on noise frames in plain Node
(about 0.9x Game Boy real time single-threaded on a WSL laptop).
Usage
import { NeuralAgent, gameboyDecoderConfig, toButtonMask } from '@flybrain/brain';
import { loadBrainDatasetFromDir } from '@flybrain/brain/node';
const dataset = await loadBrainDatasetFromDir('data/fafb-v783');
const agent = new NeuralAgent(dataset, { decoder: gameboyDecoderConfig() });
agent.warmup(firstFrame); // 2,500 ms with plasticity off, then calibrate
// every emulator frame:
const { active } = agent.tick(framebuffer, { // RGBA 160x144 by default; any size via config
rewards: [{ value: 0.5 }], // scalar rewards your game adapter detected
boot: !inGame, // relaxes Start/Select throttling on title screens
});
emulator.setButtons(toButtonMask(active));
const checkpoint = agent.exportState(); // bit-exact resume, validated on import
The lower layers (LifNetwork, RewardModulatedStdp, PopulationDecoder) are exported too for
hosts that want to run the loop themselves.
integration.md has the full per-frame loop, the fractional frame timing and the checkpoint contract.
Documentation
- Overview: the pipeline, the layer map and the design principles.
- Dataset format: artifacts, CSR layout, weight encoding, every role and its count, fingerprinting, regeneration, license.
- Model: the 1-ms LIF kernel step by step, every default constant, the retina projection, the RNG, state export and version strings.
- Plasticity: edge selection, the eligibility and reinforcement equations, statistics, topology hash and the explicit non-claims.
- Readout: scores, exclusive groups, pulse channels, the blocked-direction cooldown, the Game Boy preset table and checkpoint versions.
- Integration: what a game must provide, the Pokemon Red integration as a worked example, and a sketch of a platformer adapter.
- Limitations: what is not claimed, what is unproven, measured throughput.
- Verification: the oracle-test strategy and what each test file covers.
- Streaming plan: headless capture, Twitch, VM design and the phased plan for a 24/7 stream.
- Artifact builder: how to regenerate and verify
data/fafb-v783. - Data attribution: source, license and citations.
Provenance
The network, plasticity rule, readout, FlyWire pipeline and activity viewer were extracted from
the fly-plays-pokemon prototype so several game demos can share one core. The default
configuration reproduces that prototype's kernel bit for bit, and verbatim copies of its modules
live in packages/brain/tests/legacy/ as oracles. Built with Astra.
Licensing
The data/fafb-v783 artifacts are derived from the FlyWire FAFB public Codex v783 exports and are
licensed CC BY-NC 4.0. That is a
non-commercial license, so a commercial demo needs a different data source or separate permission.
Citations and the list of modifications are in
data/fafb-v783/ATTRIBUTION.md.
No license has been chosen for the code in this repository yet.
ROMs and save states never enter this repository.