tcai_meta_learning
Analyze reward-prediction-error variance to compute meta-learning velocity, indicating convergence or novelty spikes; optionally inject an RPE sample.
Instructions
Meta-learning state (MetaLearningModule port): learning velocity from RPE-variance dynamics. velocity>0 ⇒ converging; noveltySpike ⇒ novel/confusing regime. Optionally inject an RPE sample.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| rpe | No | Inject a reward-prediction-error sample ∈ [−1,1] |