Causal emergence predictive of final reward early in RL training across multiple algorithms, architectures, and environments.

{
  "doc": "pigozzi-levin-causally-emergent-alignment-2026.md",
  "author": null,
  "context": "Empirical result: CE measurements correlate with and predict learning performance in RL agents.",
  "enrichment": {
    "is_stale": true
  },
  "norm_label": "Causal emergence predictive of final reward early in RL training across multiple algorithms, architectures, and environments.",
  "source_url": null,
  "graphify_id": "reward_prediction_finding",
  "source_file": "pigozzi-levin-causally-emergent-alignment-2026.md",
  "imported_from": "/Users/antonborzov/Documents/Research.nosync/papers/extract_typed_out/pigozzi-levin-causally-emergent-alignment-2026/graph.json",
  "extracted_type": "finding",
  "graphify_file_type": "finding"
}