{
  "handle": "PredictiveBehaviorEmbedding",
  "ref": "PredictiveBehaviorEmbedding#c46b",
  "stub": "c46b",
  "semaId": "sema:PredictiveBehaviorEmbedding#mh:SHA-256:c46b7187c440856b539933c739aa8631852808e27fc784330102a57239d97a3e",
  "layer": "Mind",
  "category": "Inference",
  "gloss": "Train a network to predict future motion; its internal features separate behaviours it was never told about",
  "mechanism": "To find individual behavioural states without labels, train a model on a self-supervised prediction task and read its hidden layers. For ants: represent each pose as joint angles in exponential-map form (63 dimensions), take 25 observed frames at 15 fps, compress the 35-frame window (observed plus predicted) with a discrete cosine transform, and train a graph convolutional network (37 nodes, 256 hidden features, 12 residual blocks with batch norm, tanh and dropout 0.5, learnable node-to-node attention, about 2.1 M parameters) to predict the future joint angles with an L1 loss. Hook the output of the final residual block, standardise, and project with PCA: the first three components (47% of variance) already separate walking, running and antennation. A model that must forecast motion has to compress it, as in {{model_as_compression}}, and the compression aligns with behavioural state. The aim is per-agent state labels that, with network structure, let one study {{coevolution}} between individual states and collective topology.",
  "invariants": [
    "No behavioural labels are used in training; labels are only used afterwards to inspect the embedding",
    "Embeddings are taken from a fixed layer after training, then standardised before projection"
  ],
  "failureModes": [
    "Visual separation in a 3-D projection is not a classifier; the split must be validated on held-out individuals",
    "Trained on very few individuals, the features may encode identity rather than behaviour",
    "Missing joint angles from tracking failures, if imputed naively, create spurious clusters"
  ],
  "ring": 2,
  "tier": 3,
  "aliases": {
    "model_as_compression": "ModelAsCompression",
    "coevolution": "AdaptiveNetworkCoevolution"
  },
  "buildsOn": [
    {
      "handle": "ModelAsCompression",
      "type": "references",
      "alias": "model_as_compression",
      "url": "/knowledge-graph/ModelAsCompression/",
      "json": "/knowledge-graph/ModelAsCompression.json"
    },
    {
      "handle": "AdaptiveNetworkCoevolution",
      "type": "yields",
      "alias": "coevolution",
      "url": "/knowledge-graph/AdaptiveNetworkCoevolution/",
      "json": "/knowledge-graph/AdaptiveNetworkCoevolution.json"
    }
  ],
  "usedBy": [],
  "sources": {
    "research": [],
    "works": [
      {
        "kind": "thesis",
        "label": "Kevin Do, PhD thesis on ant collective behaviour (NCSU, 2026)",
        "note": "Ch. 2.3, Fig. 2.5; Ch. 5.7.3",
        "url": ""
      }
    ],
    "homePage": false
  },
  "url": "/knowledge-graph/PredictiveBehaviorEmbedding/",
  "graphUrl": "/knowledge-graph#PredictiveBehaviorEmbedding",
  "directory": "/knowledge-graph/patterns.json"
}