Rock Paper Scissors - with adaptive learning

Play against an AI that learns your patterns in real-time using ruvector SONA adaptive learning, running entirely in-browser via WebAssembly.

Loading WASM module...
0
You
0
Draw
0
AI

Rolling accuracy (last 20 rounds)

Cumulative accuracy (all time)

N-gram predictor

Accuracy-
Correct / tried0/0
Coverage-
Last prediction-

SONA LoRA

Accuracy-
Correct / tried0/0
Coverage-
Last prediction-

Combined (game)

Accuracy-
Round0

SONA engine

Learn cycles0
Trajectories buf.0
Patterns stored0
How it works: Two prediction systems work together: (1) An n-gram frequency predictor (cyan) tracks what you play after specific move sequences (window sizes 1-5). (2) The full SONA engine (purple, 227 KB WASM) records each round as a trajectory, trains two-tier LoRA weights (micro + base) with EWC++ memory preservation, and transforms context embeddings via applyLora() to predict your next move. The combined predictor (white dashed) uses n-gram when available, falls back to SONA, then random.

Tip: Try repeating a pattern (e.g. rock-rock-paper-rock-rock-paper) and watch the AI catch on. Then switch strategies and see the accuracy dip before it re-adapts.

Credits: SONA engine by ruvnet/RuVectorSONA crate.
Source on GitHub