Chris4K/quickstart-trackio-bucket
291 kB
👋 Open to Work
CK
Chris4K
AI & ML interests
Phonetics, Alpha-Beta Pruning, Bayes, Go the game, NLP, Bots
Recent Activity
reacted to theirpost with 🔥 about 5 hours ago
Can a network whose trunk is only XOR, table lookups and integer adds learn Connect Four?
Partly. I trained a GF(2⁸) integer model (8.4M params, CPU-only) on 320k games labelled by a negamax search teacher.
What it learned:
• agrees with the teacher on 59.6% of training / 41.0% of board-disjoint holdout positions
• beats a random player 85.2% of the time
What it didn't:
• loses to a 1-ply "win now / block now" rule (wins 37.2%)
• loses to 2-ply lookahead (wins 24.3%)
Agreeing with a teacher is not the same as playing well, and this release measures both.
Caveat up front: the float32 read-out is 99.6% of the parameters. The trunk is float-free; the full model isn't yet.
The repo also documents six bugs that each produced a plausible but wrong number. In one, 41% of the supervision was silently all-zero.
🎮 Play: https://huggingface.co/spaces/Chris4K/connect-four-gf256 (Space)
🧠 Model: https://huggingface.co/Chris4K/connect-four-gf256
📦 Dataset: https://huggingface.co/Chris4K/connect-four-gf256 320k games + per-ply targets, MIT
💻 Code + paper: https://github.com/Ckal/connect-four-gf256
Feedback welcome, especially: which baseline would convince you?
reacted to theirpost with 😎 1 day ago
Can a network whose trunk is only XOR, table lookups and integer adds learn Connect Four?
Partly. I trained a GF(2⁸) integer model (8.4M params, CPU-only) on 320k games labelled by a negamax search teacher.
What it learned:
• agrees with the teacher on 59.6% of training / 41.0% of board-disjoint holdout positions
• beats a random player 85.2% of the time
What it didn't:
• loses to a 1-ply "win now / block now" rule (wins 37.2%)
• loses to 2-ply lookahead (wins 24.3%)
Agreeing with a teacher is not the same as playing well, and this release measures both.
Caveat up front: the float32 read-out is 99.6% of the parameters. The trunk is float-free; the full model isn't yet.
The repo also documents six bugs that each produced a plausible but wrong number. In one, 41% of the supervision was silently all-zero.
🎮 Play: https://huggingface.co/spaces/Chris4K/connect-four-gf256 (Space)
🧠 Model: https://huggingface.co/Chris4K/connect-four-gf256
📦 Dataset: https://huggingface.co/Chris4K/connect-four-gf256 320k games + per-ply targets, MIT
💻 Code + paper: https://github.com/Ckal/connect-four-gf256
Feedback welcome, especially: which baseline would convince you?
posted an update 1 day ago
Can a network whose trunk is only XOR, table lookups and integer adds learn Connect Four?
Partly. I trained a GF(2⁸) integer model (8.4M params, CPU-only) on 320k games labelled by a negamax search teacher.
What it learned:
• agrees with the teacher on 59.6% of training / 41.0% of board-disjoint holdout positions
• beats a random player 85.2% of the time
What it didn't:
• loses to a 1-ply "win now / block now" rule (wins 37.2%)
• loses to 2-ply lookahead (wins 24.3%)
Agreeing with a teacher is not the same as playing well, and this release measures both.
Caveat up front: the float32 read-out is 99.6% of the parameters. The trunk is float-free; the full model isn't yet.
The repo also documents six bugs that each produced a plausible but wrong number. In one, 41% of the supervision was silently all-zero.
🎮 Play: https://huggingface.co/spaces/Chris4K/connect-four-gf256 (Space)
🧠 Model: https://huggingface.co/Chris4K/connect-four-gf256
📦 Dataset: https://huggingface.co/Chris4K/connect-four-gf256 320k games + per-ply targets, MIT
💻 Code + paper: https://github.com/Ckal/connect-four-gf256
Feedback welcome, especially: which baseline would convince you?
Organizations
Posts 1
Post
1,565
Can a network whose trunk is only XOR, table lookups and integer adds learn Connect Four?
Partly. I trained a GF(2⁸) integer model (8.4M params, CPU-only) on 320k games labelled by a negamax search teacher.
What it learned:
• agrees with the teacher on 59.6% of training / 41.0% of board-disjoint holdout positions
• beats a random player 85.2% of the time
What it didn't:
• loses to a 1-ply "win now / block now" rule (wins 37.2%)
• loses to 2-ply lookahead (wins 24.3%)
Agreeing with a teacher is not the same as playing well, and this release measures both.
Caveat up front: the float32 read-out is 99.6% of the parameters. The trunk is float-free; the full model isn't yet.
The repo also documents six bugs that each produced a plausible but wrong number. In one, 41% of the supervision was silently all-zero.
🎮 Play: Chris4K/connect-four-gf256 (Space)
🧠 Model: Chris4K/connect-four-gf256
📦 Dataset: Chris4K/connect-four-gf256 320k games + per-ply targets, MIT
💻 Code + paper: https://github.com/Ckal/connect-four-gf256
Feedback welcome, especially: which baseline would convince you?
Partly. I trained a GF(2⁸) integer model (8.4M params, CPU-only) on 320k games labelled by a negamax search teacher.
What it learned:
• agrees with the teacher on 59.6% of training / 41.0% of board-disjoint holdout positions
• beats a random player 85.2% of the time
What it didn't:
• loses to a 1-ply "win now / block now" rule (wins 37.2%)
• loses to 2-ply lookahead (wins 24.3%)
Agreeing with a teacher is not the same as playing well, and this release measures both.
Caveat up front: the float32 read-out is 99.6% of the parameters. The trunk is float-free; the full model isn't yet.
The repo also documents six bugs that each produced a plausible but wrong number. In one, 41% of the supervision was silently all-zero.
🎮 Play: Chris4K/connect-four-gf256 (Space)
🧠 Model: Chris4K/connect-four-gf256
📦 Dataset: Chris4K/connect-four-gf256 320k games + per-ply targets, MIT
💻 Code + paper: https://github.com/Ckal/connect-four-gf256
Feedback welcome, especially: which baseline would convince you?
spaces 137
pinned
Running
3
Layers
🐨
Explore AI system architecture with an interactive blueprint
pinned
Sleeping
1
agent-prompts — FORGE Prompt Registry
💬
Versioned prompt library, persona management, and A/B
pinned
Sleeping
2
FORGE v2 — Universal Capability Registry
⚒
Universal Capability Registry for AI Agents
pinned
Paused
1
agent-trace — FORGE Telemetry
📊
Telemetry backbone for the FORGE AI agent ecosystem
pinned
Paused
1
agent-learn — FORGE Learning Layer
🧠
Persistent Q-table, reward scoring, and RLHF store for FORGE
pinned
Paused
7
agent-loop — FORGE Self-Improvement Orchestrator
🔄
Feedback loop — trace→learn→propose→approve→deploy