{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "dd98bf31",
      "metadata": {},
      "source": [
        "# Inferring function-function edges (Tier 0)\n",
        "\n",
        "This notebook demonstrates **Tier 0** edge inference from `docs/notes/edge_inference_notes.md`: score candidate function → function edges using the correlation between per-node **activation magnitudes at layer n−1** and **gradient magnitudes at layer n**:\n",
        "\n",
        "$$\\text{score}(i \\to j) = \\mathrm{agg}_n\\, \\mathrm{corr}_b\\big(\\|z_i^{n-1}\\|,\\; \\|\\nabla_{z_j^n}\\mathcal{L}\\|\\big)$$\n",
        "\n",
        "This matches how GSNN propagates information: activations flow forward $n{-}1 \\to n$, gradients flow backward $n \\to n{-}1$.\n",
        "\n",
        "We train a GSNN on a **partial graph** (with held-out function-function edges), then use `MagnitudeEdgeInferer` to recover the missing edges post-hoc. The ground-truth graph uses the **converging-tier DAG** from the Tier 0 plan (3 inputs → 6 function nodes → 3 outputs), scaled up to **12 inputs → 24 function nodes → 12 outputs**, with **16 held-out** function-function edges removed for `G_partial`. No changes to the GSNN model are required."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "9afa6332",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "The autoreload extension is already loaded. To reload it, use:\n",
            "  %reload_ext autoreload\n"
          ]
        }
      ],
      "source": [
        "from matplotlib import pyplot as plt\n",
        "import numpy as np\n",
        "import networkx as nx\n",
        "import torch\n",
        "\n",
        "from gsnn.models.GSNN import GSNN\n",
        "from gsnn.simulate.nx2pyg import nx2pyg\n",
        "\n",
        "from gsnn.simulate.simulate import simulate\n",
        "from gsnn.optim.MagnitudeEdgeInferer import MagnitudeEdgeInferer\n",
        "\n",
        "from sklearn.metrics import r2_score, roc_auc_score, roc_curve\n",
        "\n",
        "torch.manual_seed(0)\n",
        "np.random.seed(0)\n",
        "\n",
        "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
        "\n",
        "%load_ext autoreload\n",
        "%autoreload 2"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "eb101065",
      "metadata": {},
      "source": [
        "## Build ground-truth graph G*\n",
        "\n",
        "Same **converging-tier DAG** as the Tier 0 plan, scaled 4×:\n",
        "\n",
        "| Tier | Plan (small) | This notebook |\n",
        "|------|--------------|---------------|\n",
        "| inputs → tier A | 3 → `f0…f2` | 12 → `f0…f11` |\n",
        "| tier A → tier B | overlapping pairs → `f3,f4` | → `f12…f22` |\n",
        "| tier B → tier C | `f3,f4 → f5` | `f12…f22 → f23` |\n",
        "| function → output | `f3→o0, f4→o1, f5→o2` | each tier-B node + sink → `o0…o11` |\n",
        "\n",
        "At plan scale (`n=3`): `f0→f3`, `f1→f3`, `f1→f4`, `f2→f4`, `f3→f5`, `f4→f5`. Each tier-B node `f_{n+k}` receives from `f_k` and `f_{k+1}`.\n",
        "\n",
        "**Held-out** edges (16 total) are chosen at the same structural roles, scaled up:\n",
        "- **10** tier-A → tier-B “left merge” edges `(f_k, f_{n+k})` for `k = 1, …, n−2` (includes the plan’s `f1→f4` at `n=3`)\n",
        "- **6** tier-B → sink edges on every other merge node (includes the plan’s `f3→f5` at `n=3`)\n",
        "\n",
        "These are removed when building `G_partial`; the inferrer should rank them above random non-edges."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "id": "e6433ba9",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Tier-A width: 12\n",
            "inputs: 12 | functions: 24 | outputs: 12\n",
            "Function-function edges: 33\n",
            "Held-out edges: 16 | kept in G_partial: 17\n",
            "Is DAG: True\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x700 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "def build_convergence_graph(n_tier_a=6):\n",
        "    \"\"\"Converging-tier DAG from the Tier 0 plan, parameterized by tier-A width.\n",
        "\n",
        "    Structure (n_tier_a=3 matches the plan):\n",
        "      in_k -> f_k\n",
        "      f_k, f_{k+1} -> f_{n + k}   for k = 0 .. n_tier_a-2\n",
        "      all tier-B -> f_sink\n",
        "      f_{n+k} -> o_k, f_sink -> o_{n_tier_a-1}\n",
        "    \"\"\"\n",
        "    n = n_tier_a\n",
        "    G = nx.DiGraph()\n",
        "    func_func_edges_TRUE = []\n",
        "\n",
        "    input_nodes = [f'in{i}' for i in range(n)]\n",
        "    tier_a = [f'f{i}' for i in range(n)]\n",
        "    tier_b = [f'f{n + k}' for k in range(n - 1)]\n",
        "    tier_c = [f'f{2 * n - 1}']\n",
        "    function_nodes = tier_a + tier_b + tier_c\n",
        "    output_nodes = [f'o{k}' for k in range(n)]\n",
        "\n",
        "    for i, u in enumerate(input_nodes):\n",
        "        G.add_edge(u, tier_a[i])\n",
        "\n",
        "    for k in range(n - 1):\n",
        "        b = tier_b[k]\n",
        "        for parent in (tier_a[k], tier_a[k + 1]):\n",
        "            e = (parent, b)\n",
        "            G.add_edge(*e)\n",
        "            func_func_edges_TRUE.append(e)\n",
        "\n",
        "    sink = tier_c[0]\n",
        "    for b in tier_b:\n",
        "        e = (b, sink)\n",
        "        G.add_edge(*e)\n",
        "        func_func_edges_TRUE.append(e)\n",
        "\n",
        "    for k, b in enumerate(tier_b):\n",
        "        G.add_edge(b, output_nodes[k])\n",
        "    G.add_edge(sink, output_nodes[n - 1])\n",
        "\n",
        "    # Layered layout for visualization\n",
        "    pos = {}\n",
        "    x_spread = max(n - 1, 1)\n",
        "    for i, u in enumerate(input_nodes):\n",
        "        pos[u] = (i * 2 * x_spread / max(n - 1, 1) - x_spread, 4.0)\n",
        "    for i, f in enumerate(tier_a):\n",
        "        pos[f] = (i * 2 * x_spread / max(n - 1, 1) - x_spread, 3.0)\n",
        "    for k, f in enumerate(tier_b):\n",
        "        pos[f] = ((k + 0.5) * 2 * x_spread / max(n - 2, 1) - x_spread, 2.0)\n",
        "    pos[sink] = (0.0, 1.0)\n",
        "    for k, o in enumerate(output_nodes):\n",
        "        pos[o] = (k * 2 * x_spread / max(n - 1, 1) - x_spread, 0.0)\n",
        "\n",
        "    return G, pos, input_nodes, function_nodes, output_nodes, func_func_edges_TRUE\n",
        "\n",
        "\n",
        "def default_held_out_edges(n_tier_a, b2sink_stride=2):\n",
        "    \"\"\"Held-out ff edges at scaled plan analogues.\n",
        "\n",
        "    A→B: (f_k, f_{n+k}) for k=1..n-2 — left parent of each merge (f1→f4 at n=3).\n",
        "    B→sink: (f_{n+k}, f_{2n-1}) for k=0, stride, 2*stride, ... (f3→f5 at n=3).\n",
        "    \"\"\"\n",
        "    n = n_tier_a\n",
        "    sink = f'f{2 * n - 1}'\n",
        "    held = [(f'f{k}', f'f{n + k}') for k in range(1, n - 1)]\n",
        "    held += [(f'f{n + k}', sink) for k in range(0, n - 1, b2sink_stride)]\n",
        "    return held\n",
        "\n",
        "\n",
        "N_TIER_A = 12\n",
        "G, pos, input_nodes, function_nodes, output_nodes, func_func_edges_TRUE = build_convergence_graph(\n",
        "    n_tier_a=N_TIER_A,\n",
        ")\n",
        "N_FUNC = len(function_nodes)\n",
        "\n",
        "HELD_OUT_EDGES = default_held_out_edges(N_TIER_A, b2sink_stride=2)\n",
        "held_out_set = set(HELD_OUT_EDGES)\n",
        "kept_ff_set = set(func_func_edges_TRUE) - held_out_set\n",
        "\n",
        "assert all(G.has_edge(*e) for e in HELD_OUT_EDGES)\n",
        "assert len(held_out_set & kept_ff_set) == 0\n",
        "\n",
        "print(f'Tier-A width: {N_TIER_A}')\n",
        "print(f'inputs: {len(input_nodes)} | functions: {N_FUNC} | outputs: {len(output_nodes)}')\n",
        "print(f'Function-function edges: {len(func_func_edges_TRUE)}')\n",
        "print(f'Held-out edges: {len(HELD_OUT_EDGES)} | kept in G_partial: {len(kept_ff_set)}')\n",
        "print(f'Is DAG: {nx.is_directed_acyclic_graph(G)}')\n",
        "\n",
        "node_color = []\n",
        "for n in G.nodes:\n",
        "    if n in input_nodes:\n",
        "        node_color.append('#9ecae1')\n",
        "    elif n in output_nodes:\n",
        "        node_color.append('#fdae6b')\n",
        "    else:\n",
        "        node_color.append('#a1d99b')\n",
        "\n",
        "edge_colors = ['C3' if e in held_out_set else 'gray' for e in G.edges]\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(16, 7))\n",
        "nx.draw_networkx(\n",
        "    G, pos, ax=ax, with_labels=True, node_color=node_color, node_size=500,\n",
        "    font_size=6, arrowsize=10, edge_color=edge_colors,\n",
        ")\n",
        "ax.set_title(f'True graph G*: converging DAG (n={N_TIER_A}; held-out edges in red)')\n",
        "ax.set_axis_off()\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0ec38d72",
      "metadata": {},
      "source": [
        "## Simulate data\n",
        "\n",
        "Use the default linear-Gaussian generative process from `simulate` (as in the Tier 0 plan) so every edge carries signal."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e01babc7",
      "metadata": {},
      "outputs": [],
      "source": [
        "N_TRAIN = 2000\n",
        "N_TEST = 500\n",
        "\n",
        "x_train, x_test, y_train, y_test = simulate(\n",
        "    G,\n",
        "    n_train=N_TRAIN,\n",
        "    n_test=N_TEST,\n",
        "    input_nodes=input_nodes,\n",
        "    output_nodes=output_nodes,\n",
        "    noise_scale=0.15,\n",
        "    special_functions=None,\n",
        ")\n",
        "\n",
        "x_train = torch.tensor(x_train, dtype=torch.float32).to(device)\n",
        "x_test = torch.tensor(x_test, dtype=torch.float32).to(device)\n",
        "y_train = torch.tensor(y_train, dtype=torch.float32).to(device)\n",
        "y_test = torch.tensor(y_test, dtype=torch.float32).to(device)\n",
        "\n",
        "y_mu = y_train.mean(0)\n",
        "y_std = y_train.std(0)\n",
        "y_train = (y_train - y_mu) / (y_std + 1e-8)\n",
        "y_test = (y_test - y_mu) / (y_std + 1e-8)\n",
        "\n",
        "print('train:', x_train.shape, y_train.shape)\n",
        "print('test:', x_test.shape, y_test.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "c0130065",
      "metadata": {},
      "source": [
        "## Build partial graph G_partial\n",
        "\n",
        "Remove the designated held-out function-function edges. The model trains on the remaining graph; we try to recover those edges post-hoc."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "id": "554f7419",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Held-out function-function edges: 16\n",
            "Kept function-function edges: 17\n",
            "Non-edges (true negatives): 519\n",
            "G_partial edges: 41\n",
            "Held-out: [('f1', 'f13'), ('f2', 'f14'), ('f3', 'f15'), ('f4', 'f16'), ('f5', 'f17'), ('f6', 'f18'), ('f7', 'f19'), ('f8', 'f20'), ('f9', 'f21'), ('f10', 'f22'), ('f12', 'f23'), ('f14', 'f23'), ('f16', 'f23'), ('f18', 'f23'), ('f20', 'f23'), ('f22', 'f23')]\n"
          ]
        }
      ],
      "source": [
        "held_out_edges = HELD_OUT_EDGES\n",
        "held_out_set = set(held_out_edges)\n",
        "\n",
        "kept_edges = [e for e in func_func_edges_TRUE if e not in held_out_set]\n",
        "true_ff_set = set(func_func_edges_TRUE)\n",
        "\n",
        "G_partial = G.copy()\n",
        "G_partial.remove_edges_from(held_out_edges)\n",
        "data = nx2pyg(G_partial, input_nodes, function_nodes, output_nodes)\n",
        "\n",
        "n_non_edges = N_FUNC * (N_FUNC - 1) - len(true_ff_set)\n",
        "\n",
        "print(f'Held-out function-function edges: {len(held_out_edges)}')\n",
        "print(f'Kept function-function edges: {len(kept_edges)}')\n",
        "print(f'Non-edges (true negatives): {n_non_edges}')\n",
        "print(f'G_partial edges: {G_partial.number_of_edges()}')\n",
        "print('Held-out:', held_out_edges)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "350ac0f2",
      "metadata": {},
      "source": [
        "## Train GSNN on G_partial"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "id": "e662f88b",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "n params 6924\n",
            "epoch  1/30 | train loss: 0.6953\n",
            "epoch  5/30 | train loss: 0.4929\n",
            "epoch 10/30 | train loss: 0.4903\n",
            "epoch 15/30 | train loss: 0.4866\n",
            "epoch 20/30 | train loss: 0.4859\n",
            "epoch 25/30 | train loss: 0.4857\n",
            "epoch 30/30 | train loss: 0.4854\n",
            "test loss: 0.4801 | test r2: 0.516\n"
          ]
        }
      ],
      "source": [
        "BATCH_SIZE = 64\n",
        "\n",
        "model_kwargs = {\n",
        "    'channels': 8,\n",
        "    'layers': 6,\n",
        "    'share_layers': False,\n",
        "    'bias': True,\n",
        "    'add_function_self_edges': True,\n",
        "    'norm': 'groupbatch',\n",
        "    'dropout': 0.,\n",
        "    'nonlin': torch.nn.ELU,\n",
        "    'node_mlp': False,\n",
        "    'checkpoint': False,\n",
        "}\n",
        "\n",
        "model = GSNN(\n",
        "    data.edge_index_dict,\n",
        "    data.node_names_dict,\n",
        "    **model_kwargs,\n",
        ").to(device)\n",
        "\n",
        "print('n params', sum(p.numel() for p in model.parameters()))\n",
        "\n",
        "train_loader = torch.utils.data.DataLoader(\n",
        "    torch.utils.data.TensorDataset(x_train, y_train),\n",
        "    batch_size=BATCH_SIZE,\n",
        "    shuffle=True,\n",
        "    drop_last=True,  # fixed batch size for GSNN batch-param cache\n",
        ")\n",
        "\n",
        "optim = torch.optim.AdamW(model.parameters(), lr=1e-3, weight_decay=0)\n",
        "crit = torch.nn.MSELoss()\n",
        "\n",
        "n_epochs = 30\n",
        "for epoch in range(n_epochs):\n",
        "    epoch_losses = []\n",
        "    for x_batch, y_batch in train_loader:\n",
        "        optim.zero_grad()\n",
        "        yhat = model(x_batch)\n",
        "        loss = crit(yhat, y_batch)\n",
        "        loss.backward()\n",
        "        optim.step()\n",
        "        epoch_losses.append(loss.item())\n",
        "\n",
        "    if epoch == 0 or (epoch + 1) % 5 == 0 or epoch == n_epochs - 1:\n",
        "        print(f'epoch {epoch + 1:2d}/{n_epochs} | train loss: {np.mean(epoch_losses):.4f}')\n",
        "\n",
        "model.eval()\n",
        "with torch.inference_mode():\n",
        "    yhat_test = model(x_test)\n",
        "loss_test = crit(y_test, yhat_test)\n",
        "r2_test = r2_score(y_test.detach().cpu().numpy(), yhat_test.detach().cpu().numpy())\n",
        "print(f'test loss: {loss_test.item():.4f} | test r2: {r2_test:.3f}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "36865a92",
      "metadata": {},
      "source": [
        "## Run MagnitudeEdgeInferer\n",
        "\n",
        "Post-hoc pass over the training data: forward + backward per batch, accumulate streaming correlation statistics."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "id": "bb21d611",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[batch 31/31] n=1984\n",
            "Processed 1984 samples\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>src_func</th>\n",
              "      <th>dst_func</th>\n",
              "      <th>src_idx</th>\n",
              "      <th>dst_idx</th>\n",
              "      <th>corr</th>\n",
              "      <th>has_edge</th>\n",
              "      <th>corr_a0_g1</th>\n",
              "      <th>corr_a1_g2</th>\n",
              "      <th>corr_a2_g3</th>\n",
              "      <th>corr_a3_g4</th>\n",
              "      <th>corr_a4_g5</th>\n",
              "      <th>p_value</th>\n",
              "      <th>q_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>f2</td>\n",
              "      <td>f14</td>\n",
              "      <td>2</td>\n",
              "      <td>14</td>\n",
              "      <td>0.895113</td>\n",
              "      <td>False</td>\n",
              "      <td>0.895113</td>\n",
              "      <td>0.853276</td>\n",
              "      <td>0.779297</td>\n",
              "      <td>0.788583</td>\n",
              "      <td>0.848483</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>f6</td>\n",
              "      <td>f18</td>\n",
              "      <td>6</td>\n",
              "      <td>18</td>\n",
              "      <td>0.886441</td>\n",
              "      <td>False</td>\n",
              "      <td>0.874846</td>\n",
              "      <td>0.886441</td>\n",
              "      <td>0.805949</td>\n",
              "      <td>0.798208</td>\n",
              "      <td>0.742110</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>f15</td>\n",
              "      <td>f16</td>\n",
              "      <td>15</td>\n",
              "      <td>16</td>\n",
              "      <td>0.879697</td>\n",
              "      <td>False</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.600171</td>\n",
              "      <td>0.819732</td>\n",
              "      <td>0.879697</td>\n",
              "      <td>0.863181</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>f4</td>\n",
              "      <td>f16</td>\n",
              "      <td>4</td>\n",
              "      <td>16</td>\n",
              "      <td>0.876376</td>\n",
              "      <td>False</td>\n",
              "      <td>0.876376</td>\n",
              "      <td>0.787496</td>\n",
              "      <td>0.855820</td>\n",
              "      <td>0.811545</td>\n",
              "      <td>0.868536</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>f8</td>\n",
              "      <td>f20</td>\n",
              "      <td>8</td>\n",
              "      <td>20</td>\n",
              "      <td>0.870422</td>\n",
              "      <td>False</td>\n",
              "      <td>0.870422</td>\n",
              "      <td>0.807645</td>\n",
              "      <td>0.820510</td>\n",
              "      <td>0.784994</td>\n",
              "      <td>0.771802</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>f13</td>\n",
              "      <td>f14</td>\n",
              "      <td>13</td>\n",
              "      <td>14</td>\n",
              "      <td>0.861188</td>\n",
              "      <td>False</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.861188</td>\n",
              "      <td>0.744196</td>\n",
              "      <td>0.676358</td>\n",
              "      <td>0.722777</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>f20</td>\n",
              "      <td>f21</td>\n",
              "      <td>20</td>\n",
              "      <td>21</td>\n",
              "      <td>0.859313</td>\n",
              "      <td>False</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.420173</td>\n",
              "      <td>0.695480</td>\n",
              "      <td>0.859313</td>\n",
              "      <td>0.760776</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>f3</td>\n",
              "      <td>f15</td>\n",
              "      <td>3</td>\n",
              "      <td>15</td>\n",
              "      <td>0.853414</td>\n",
              "      <td>False</td>\n",
              "      <td>0.834993</td>\n",
              "      <td>0.853414</td>\n",
              "      <td>0.721508</td>\n",
              "      <td>0.778630</td>\n",
              "      <td>0.762166</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>f17</td>\n",
              "      <td>f18</td>\n",
              "      <td>17</td>\n",
              "      <td>18</td>\n",
              "      <td>0.852540</td>\n",
              "      <td>False</td>\n",
              "      <td>0.000772</td>\n",
              "      <td>0.784241</td>\n",
              "      <td>0.788614</td>\n",
              "      <td>0.852540</td>\n",
              "      <td>0.764563</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>f10</td>\n",
              "      <td>f22</td>\n",
              "      <td>10</td>\n",
              "      <td>22</td>\n",
              "      <td>0.850839</td>\n",
              "      <td>False</td>\n",
              "      <td>0.850839</td>\n",
              "      <td>0.825376</td>\n",
              "      <td>0.835415</td>\n",
              "      <td>0.748628</td>\n",
              "      <td>0.709853</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  src_func dst_func  src_idx  dst_idx      corr  has_edge  corr_a0_g1  \\\n",
              "0       f2      f14        2       14  0.895113     False    0.895113   \n",
              "1       f6      f18        6       18  0.886441     False    0.874846   \n",
              "2      f15      f16       15       16  0.879697     False         NaN   \n",
              "3       f4      f16        4       16  0.876376     False    0.876376   \n",
              "4       f8      f20        8       20  0.870422     False    0.870422   \n",
              "5      f13      f14       13       14  0.861188     False         NaN   \n",
              "6      f20      f21       20       21  0.859313     False         NaN   \n",
              "7       f3      f15        3       15  0.853414     False    0.834993   \n",
              "8      f17      f18       17       18  0.852540     False    0.000772   \n",
              "9      f10      f22       10       22  0.850839     False    0.850839   \n",
              "\n",
              "   corr_a1_g2  corr_a2_g3  corr_a3_g4  corr_a4_g5  p_value  q_value  \n",
              "0    0.853276    0.779297    0.788583    0.848483      0.0      0.0  \n",
              "1    0.886441    0.805949    0.798208    0.742110      0.0      0.0  \n",
              "2    0.600171    0.819732    0.879697    0.863181      0.0      0.0  \n",
              "3    0.787496    0.855820    0.811545    0.868536      0.0      0.0  \n",
              "4    0.807645    0.820510    0.784994    0.771802      0.0      0.0  \n",
              "5    0.861188    0.744196    0.676358    0.722777      0.0      0.0  \n",
              "6    0.420173    0.695480    0.859313    0.760776      0.0      0.0  \n",
              "7    0.853414    0.721508    0.778630    0.762166      0.0      0.0  \n",
              "8    0.784241    0.788614    0.852540    0.764563      0.0      0.0  \n",
              "9    0.825376    0.835415    0.748628    0.709853      0.0      0.0  "
            ]
          },
          "execution_count": 40,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "MEI = MagnitudeEdgeInferer(model, data, reduction='l1')\n",
        "\n",
        "# Edge inferrer runs forward+backward per batch; reuse the same fixed-size loader.\n",
        "infer_loader = torch.utils.data.DataLoader(\n",
        "    torch.utils.data.TensorDataset(x_train, y_train),\n",
        "    batch_size=BATCH_SIZE,\n",
        "    shuffle=False,\n",
        "    drop_last=True,\n",
        ")\n",
        "\n",
        "n_samples = MEI.fit(infer_loader, crit=crit, device=device, verbose=True)\n",
        "res = MEI.evaluate(layer_agg='max')\n",
        "\n",
        "print(f'Processed {n_samples} samples')\n",
        "res.head(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "id": "f7caff79",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Top 10 scored pairs:\n",
            "  src_func dst_func      corr edge_category  has_edge  q_value\n",
            "0       f2      f14  0.895113      held_out     False      0.0\n",
            "1       f6      f18  0.886441      held_out     False      0.0\n",
            "2      f15      f16  0.879697      non_edge     False      0.0\n",
            "3       f4      f16  0.876376      held_out     False      0.0\n",
            "4       f8      f20  0.870422      held_out     False      0.0\n",
            "5      f13      f14  0.861188      non_edge     False      0.0\n",
            "6      f20      f21  0.859313      non_edge     False      0.0\n",
            "7       f3      f15  0.853414      held_out     False      0.0\n",
            "8      f17      f18  0.852540      non_edge     False      0.0\n",
            "9      f10      f22  0.850839      held_out     False      0.0\n",
            "\n",
            "Category summary:\n",
            "               count      mean    median       std\n",
            "edge_category                                     \n",
            "held_out          16  0.576020  0.822944  0.367172\n",
            "in_graph          17 -0.136474 -0.167437  0.097528\n",
            "non_edge         519  0.045526 -0.010472  0.199375\n",
            "\n",
            "Precision / recall in top-k ranked pairs:\n",
            "  top- 10: precision=0.6000, recall=0.3750\n",
            "  top- 25: precision=0.3600, recall=0.5625\n",
            "  top- 50: precision=0.2200, recall=0.6875\n",
            "  top-100: precision=0.1500, recall=0.9375\n"
          ]
        }
      ],
      "source": [
        "kept_set = set(kept_edges)\n",
        "\n",
        "def edge_category(row):\n",
        "    pair = (row['src_func'], row['dst_func'])\n",
        "    if pair in held_out_set:\n",
        "        return 'held_out'\n",
        "    if pair in kept_set:\n",
        "        return 'in_graph'\n",
        "    return 'non_edge'\n",
        "\n",
        "res = res.assign(edge_category=res.apply(edge_category, axis=1))\n",
        "\n",
        "print('Top 10 scored pairs:')\n",
        "print(res[['src_func', 'dst_func', 'corr', 'edge_category', 'has_edge', 'q_value']].head(10).to_string())\n",
        "\n",
        "print('\\nCategory summary:')\n",
        "print(res.groupby('edge_category')['corr'].agg(['count', 'mean', 'median', 'std']).to_string())\n",
        "\n",
        "print('\\nPrecision / recall in top-k ranked pairs:')\n",
        "for k in [10, 25, 50, 100]:\n",
        "    if k > len(res):\n",
        "        continue\n",
        "    topk = res.head(k)\n",
        "    prec = (topk['edge_category'] == 'held_out').mean()\n",
        "    recall = (topk['edge_category'] == 'held_out').sum() / len(held_out_edges)\n",
        "    print(f'  top-{k:3d}: precision={prec:.4f}, recall={recall:.4f}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "e2d12cb5",
      "metadata": {},
      "source": [
        "## Validation plots\n",
        "\n",
        "Expectation:\n",
        "- **held_out** edges (missing from G_partial) should score higher than **non_edge** pairs\n",
        "- **in_graph** edges may score near zero or negative due to **gradient absorption** at equilibrium (see discussion below)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "id": "c975adb0",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x400 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, axes = plt.subplots(1, 2, figsize=(10, 4))\n",
        "\n",
        "# Violin plot by category\n",
        "categories = ['held_out', 'in_graph', 'non_edge']\n",
        "data_plot = [res.loc[res.edge_category == c, 'corr'].dropna().values for c in categories]\n",
        "parts = axes[0].violinplot(data_plot, showmeans=True, showmedians=True)\n",
        "axes[0].set_xticks(range(1, len(categories) + 1))\n",
        "axes[0].set_xticklabels(categories)\n",
        "axes[0].set_ylabel('corr(|z_i|, |grad z_j|)')\n",
        "axes[0].set_title('Score distribution by edge category')\n",
        "axes[0].axhline(0, color='k', lw=0.5, alpha=0.5)\n",
        "\n",
        "# ROC: held_out (positive) vs non_edge (negative)\n",
        "pos = res[res.edge_category == 'held_out']['corr'].dropna().values\n",
        "neg = res[res.edge_category == 'non_edge']['corr'].dropna().values\n",
        "y_true = np.concatenate([np.ones(len(pos)), np.zeros(len(neg))])\n",
        "y_score = np.concatenate([pos, neg])\n",
        "\n",
        "if len(np.unique(y_true)) == 2:\n",
        "    auc = roc_auc_score(y_true, y_score)\n",
        "    fpr, tpr, _ = roc_curve(y_true, y_score)\n",
        "    axes[1].plot(fpr, tpr, label=f'AUC = {auc:.3f}')\n",
        "    axes[1].plot([0, 1], [0, 1], 'k--', alpha=0.5)\n",
        "    axes[1].set_xlabel('FPR')\n",
        "    axes[1].set_ylabel('TPR')\n",
        "    axes[1].set_title('ROC: held_out vs non_edge')\n",
        "    axes[1].legend()\n",
        "else:\n",
        "    axes[1].text(0.5, 0.5, 'Insufficient classes for ROC', ha='center')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "id": "f1c075b2",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x400 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Score histogram by category\n",
        "fig, ax = plt.subplots(figsize=(8, 4))\n",
        "for cat, color in zip(['held_out', 'in_graph', 'non_edge'], ['C2', 'C0', 'C1']):\n",
        "    vals = res.loc[res.edge_category == cat, 'corr'].dropna().values\n",
        "    ax.hist(vals, bins=30, alpha=0.45, label=f'{cat} (n={len(vals)})', color=color, density=True)\n",
        "ax.axvline(0, color='k', lw=0.5, alpha=0.5)\n",
        "ax.set_xlabel('corr(|z_i|, |grad z_j|)')\n",
        "ax.set_ylabel('density')\n",
        "ax.set_title(f'Score distributions ({N_FUNC}-node graph)')\n",
        "ax.legend()\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "56ff3ed5",
      "metadata": {},
      "source": [
        "## Partial correlation: controlling for kept parents\n",
        "\n",
        "The plain correlation score is inflated by **transitive paths** and **common upstream causes**: if `i → k → j` already exists in `G_partial`, then `|z_i^{n-1}|` correlates with `|z_k^{n-1}|` (shared upstream drive), which drives `|∇z_j^n|`, so `i → j` looks high even when no direct edge exists.\n",
        "\n",
        "`MagnitudeEdgeInferer` accumulates one extra statistic during `fit` — the per-layer activation Gram matrix `sum_xx[p]` — which lets us evaluate **partial correlation** at score time, conditioning on the kept parents $S_j = \\mathrm{parents}_{G_\\mathrm{partial}}(j)$:\n",
        "\n",
        "$$s_n(i \\to j) = \\mathrm{pcorr}_b\\big(\\|z_i^{n-1}\\|,\\ \\|\\nabla_{z_j^n}\\mathcal{L}\\| \\;\\big|\\; \\{\\|z_k^{n-1}\\| : k \\in S_j\\}\\big).$$\n",
        "\n",
        "Computed in closed form via Schur complement on the streamed sufficient statistics — no extra forward/backward pass. Entries with `i ∈ S_j` (kept edges) are returned as NaN since the source is part of the conditioning set; in-graph edges are therefore correctly absent from the ranked output. Fisher-z p-values use `df = n - 3 - |S_j|`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "id": "f20fc712",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Top 10 partial-correlation pairs:\n",
            "  src_func dst_func      corr edge_category  has_edge  n_cond  q_value\n",
            "0       f2      f14  0.895741      held_out     False       1      0.0\n",
            "1       f6      f18  0.888803      held_out     False       1      0.0\n",
            "2       f4      f16  0.881591      held_out     False       1      0.0\n",
            "3      f15      f16  0.880470      non_edge     False       1      0.0\n",
            "4       f8      f20  0.870521      held_out     False       1      0.0\n",
            "5      f17      f18  0.863198      non_edge     False       1      0.0\n",
            "6      f13      f14  0.863124      non_edge     False       1      0.0\n",
            "7       f3      f15  0.862425      held_out     False       1      0.0\n",
            "8      f20      f21  0.860680      non_edge     False       1      0.0\n",
            "9       f5      f17  0.859305      held_out     False       1      0.0\n",
            "\n",
            "Category summary (partial corr):\n",
            "               count      mean   median       std\n",
            "edge_category                                    \n",
            "held_out          16  0.579503  0.82903  0.369199\n",
            "in_graph           0       NaN      NaN       NaN\n",
            "non_edge         519  0.046005 -0.00940  0.199791\n",
            "\n",
            "Precision / recall in top-k (partial corr):\n",
            "  top- 10: precision=0.6000, recall=0.3750\n",
            "  top- 25: precision=0.3600, recall=0.5625\n",
            "  top- 50: precision=0.2200, recall=0.6875\n",
            "  top-100: precision=0.1500, recall=0.9375\n"
          ]
        }
      ],
      "source": [
        "res_partial = MEI.evaluate(layer_agg='max', score='partial')\n",
        "res_partial = res_partial.assign(edge_category=res_partial.apply(edge_category, axis=1))\n",
        "\n",
        "print('Top 10 partial-correlation pairs:')\n",
        "cols = ['src_func', 'dst_func', 'corr', 'edge_category', 'has_edge', 'n_cond', 'q_value']\n",
        "print(res_partial[cols].head(10).to_string())\n",
        "\n",
        "print('\\nCategory summary (partial corr):')\n",
        "print(res_partial.groupby('edge_category')['corr'].agg(['count', 'mean', 'median', 'std']).to_string())\n",
        "\n",
        "print('\\nPrecision / recall in top-k (partial corr):')\n",
        "for k in [10, 25, 50, 100]:\n",
        "    if k > len(res_partial):\n",
        "        continue\n",
        "    topk = res_partial.head(k)\n",
        "    prec = (topk['edge_category'] == 'held_out').mean()\n",
        "    recall = (topk['edge_category'] == 'held_out').sum() / len(held_out_edges)\n",
        "    print(f'  top-{k:3d}: precision={prec:.4f}, recall={recall:.4f}')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "id": "9ea2ab45",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1100x400 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "corr    AUC: 0.943\n",
            "partial AUC: 0.942\n"
          ]
        }
      ],
      "source": [
        "def _roc(df, ax, label, color):\n",
        "    pos = df[df.edge_category == 'held_out']['corr'].dropna().values\n",
        "    neg = df[df.edge_category == 'non_edge']['corr'].dropna().values\n",
        "    if len(pos) == 0 or len(neg) == 0:\n",
        "        return None\n",
        "    y_true = np.concatenate([np.ones(len(pos)), np.zeros(len(neg))])\n",
        "    y_score = np.concatenate([pos, neg])\n",
        "    auc = roc_auc_score(y_true, y_score)\n",
        "    fpr, tpr, _ = roc_curve(y_true, y_score)\n",
        "    ax.plot(fpr, tpr, color=color, label=f'{label} (AUC = {auc:.3f})')\n",
        "    return auc\n",
        "\n",
        "fig, axes = plt.subplots(1, 2, figsize=(11, 4))\n",
        "\n",
        "auc_corr = _roc(res, axes[0], 'corr', 'C0')\n",
        "auc_part = _roc(res_partial, axes[0], 'partial', 'C1')\n",
        "axes[0].plot([0, 1], [0, 1], 'k--', alpha=0.5)\n",
        "axes[0].set_xlabel('FPR')\n",
        "axes[0].set_ylabel('TPR')\n",
        "axes[0].set_title('ROC: held_out vs non_edge')\n",
        "axes[0].legend()\n",
        "\n",
        "cats = ['held_out', 'non_edge']\n",
        "data_corr = [res.loc[res.edge_category == c, 'corr'].dropna().values for c in cats]\n",
        "data_part = [res_partial.loc[res_partial.edge_category == c, 'corr'].dropna().values for c in cats]\n",
        "\n",
        "vp1 = axes[1].violinplot(data_corr, positions=[1, 4], widths=0.8, showmeans=True)\n",
        "vp2 = axes[1].violinplot(data_part, positions=[2, 5], widths=0.8, showmeans=True)\n",
        "for b in vp1['bodies']:\n",
        "    b.set_facecolor('C0'); b.set_alpha(0.5)\n",
        "for b in vp2['bodies']:\n",
        "    b.set_facecolor('C1'); b.set_alpha(0.5)\n",
        "axes[1].set_xticks([1.5, 4.5])\n",
        "axes[1].set_xticklabels(cats)\n",
        "axes[1].axhline(0, color='k', lw=0.5, alpha=0.5)\n",
        "axes[1].set_ylabel('score')\n",
        "axes[1].set_title('corr (blue) vs partial (orange) by category')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "\n",
        "print(f'corr    AUC: {auc_corr:.3f}')\n",
        "print(f'partial AUC: {auc_part:.3f}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "e737e7f3",
      "metadata": {},
      "source": [
        "## Within-target ranking (MRR, top@k)\n",
        "\n",
        "Global ROC AUC pools all `(src, dst)` pairs and treats every non-edge as a negative. That can hide a more operational question: **for each target `j` with a missing parent, is the correct source ranked first among all candidates for `j`?**\n",
        "\n",
        "`MagnitudeEdgeInferer.evaluate_target_ranking` ranks candidate sources per target (descending score) and reports:\n",
        "\n",
        "- **MRR** (mean reciprocal rank): average of `1 / rank` over held-out edges\n",
        "- **top@k**: fraction of held-out edges whose true source is in the top `k` for that target\n",
        "\n",
        "This is stricter than global top-k precision — each target gets its own mini leaderboard."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "id": "e3acd023",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== corr ===\n",
            "  MRR: 0.636  (n=16)\n",
            "  top@1: 0.500\n",
            "  top@3: 0.688\n",
            "  top@5: 0.750\n",
            "\n",
            "dst_func src_func    score  rank  n_candidates  reciprocal_rank  top@1  top@3  top@5\n",
            "     f13       f1 0.802505     1            23         1.000000   True   True   True\n",
            "     f14       f2 0.895113     1            23         1.000000   True   True   True\n",
            "     f15       f3 0.853414     1            23         1.000000   True   True   True\n",
            "     f17       f5 0.848134     1            23         1.000000   True   True   True\n",
            "     f18       f6 0.886441     1            23         1.000000   True   True   True\n",
            "     f20       f8 0.870422     1            23         1.000000   True   True   True\n",
            "     f22      f10 0.850839     1            23         1.000000   True   True   True\n",
            "     f23      f14 0.200330     1            23         1.000000   True   True   True\n",
            "     f16       f4 0.876376     2            23         0.500000  False   True   True\n",
            "     f19       f7 0.843384     2            23         0.500000  False   True   True\n",
            "     f21       f9 0.765648     2            23         0.500000  False   True   True\n",
            "     f23      f12 0.142956     4            23         0.250000  False  False   True\n",
            "     f23      f18 0.134085     7            23         0.142857  False  False  False\n",
            "     f23      f20 0.131269     9            23         0.111111  False  False  False\n",
            "     f23      f16 0.095078    10            23         0.100000  False  False  False\n",
            "     f23      f22 0.020319    13            23         0.076923  False  False  False\n",
            "\n",
            "=== partial ===\n",
            "  MRR: 0.668  (n=16)\n",
            "  top@1: 0.562\n",
            "  top@3: 0.688\n",
            "  top@5: 0.750\n",
            "\n",
            "dst_func src_func    score  rank  n_candidates  reciprocal_rank  top@1  top@3  top@5\n",
            "     f13       f1 0.811021     1            22         1.000000   True   True   True\n",
            "     f14       f2 0.895741     1            22         1.000000   True   True   True\n",
            "     f15       f3 0.862425     1            22         1.000000   True   True   True\n",
            "     f16       f4 0.881591     1            22         1.000000   True   True   True\n",
            "     f17       f5 0.859305     1            22         1.000000   True   True   True\n",
            "     f18       f6 0.888803     1            22         1.000000   True   True   True\n",
            "     f20       f8 0.870521     1            22         1.000000   True   True   True\n",
            "     f22      f10 0.855103     1            22         1.000000   True   True   True\n",
            "     f23      f14 0.201640     1            18         1.000000   True   True   True\n",
            "     f19       f7 0.847039     2            22         0.500000  False   True   True\n",
            "     f21       f9 0.771332     2            22         0.500000  False   True   True\n",
            "     f23      f12 0.143148     4            18         0.250000  False  False   True\n",
            "     f23      f18 0.135820     7            18         0.142857  False  False  False\n",
            "     f23      f20 0.130251     9            18         0.111111  False  False  False\n",
            "     f23      f16 0.099992    10            18         0.100000  False  False  False\n",
            "     f23      f22 0.018314    12            18         0.083333  False  False  False\n",
            "\n"
          ]
        }
      ],
      "source": [
        "def print_target_ranking(label, res_df, positive_edges, top_k=(1, 3, 5)):\n",
        "    detail, summary = MagnitudeEdgeInferer.evaluate_target_ranking(\n",
        "        res_df, positive_edges=positive_edges, top_k=top_k,\n",
        "    )\n",
        "    print(f'=== {label} ===')\n",
        "    print(f\"  MRR: {summary['mrr']:.3f}  (n={int(summary['n_positives'])})\")\n",
        "    for k in top_k:\n",
        "        print(f\"  top@{k}: {summary[f'top@{k}']:.3f}\")\n",
        "    print()\n",
        "    cols = ['dst_func', 'src_func', 'score', 'rank', 'n_candidates', 'reciprocal_rank'] + [f'top@{k}' for k in top_k]\n",
        "    print(detail.sort_values(['rank', 'dst_func'])[cols].to_string(index=False))\n",
        "    print()\n",
        "    return detail, summary\n",
        "\n",
        "rank_corr_detail, rank_corr_summary = print_target_ranking('corr', res, held_out_edges)\n",
        "rank_part_detail, rank_part_summary = print_target_ranking('partial', res_partial, held_out_edges)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "000c0015",
      "metadata": {},
      "source": [
        "## Discussion\n",
        "\n",
        "This notebook implements **Tier 0** from `docs/notes/edge_inference_notes.md` (section 4): a parameter-free score\n",
        "\n",
        "$$\\mathrm{corr}_b\\big(\\|z_i^{n-1}\\|,\\; \\|\\nabla_{z_j^n}\\mathcal{L}\\|\\big)$$\n",
        "\n",
        "computed post-hoc over a trained GSNN. Scores are aggregated across adjacent layer pairs $(n{-}1, n)$ with `layer_agg='mean'` (or `'max'`). By default, magnitudes are taken from **pre-norm activations** (post-``lin_in``, before the ResBlock norm layer) via `ResBlock._last_pre_norm_activation`, with `retain_grad()` for the backward pass.\n",
        "\n",
        "### Adjacent-layer pairing\n",
        "\n",
        "We correlate **activation magnitudes at layer $n{-}1$** with **gradient magnitudes at layer $n$**, not both at the same layer. That matches how GSNN propagates information: forward pass sends activations $z^{n-1} \\to z^n$ via message passing; backward pass sends gradients $\\nabla z^n \\to \\nabla z^{n-1}$. A missing edge $i \\to j$ should show up when upstream activity at $i$ in layer $n{-}1$ co-moves with downstream loss sensitivity at $j$ in layer $n$.\n",
        "\n",
        "Per-pair columns in the output are named `corr_a0_g1`, `corr_a1_g2`, … (`a` = activation layer, `g` = gradient layer).\n",
        "\n",
        "### Gradient absorption (sections 4.5 and 7.1)\n",
        "\n",
        "At convergence, if edge `i → j` already exists in `G_partial`, the model is locally optimal w.r.t. that input: when `|z_i|` is large, `|∇_{z_j} L|` is *small*. **In-graph edges therefore tend to score near zero or negative**, not high. This is expected and informative — the score highlights *missing-but-useful* edges, not edges the model already uses.\n",
        "\n",
        "### Partial correlation (`score='partial'`)\n",
        "\n",
        "The plain Pearson score above suffers from two structural confounds: **transitive paths** (`i → k → j` makes `i → j` look high through `k`) and **common upstream causes** (a shared driver `u → i, u → j` inflates the score of an absent edge). Partial correlation conditioning on the **kept parents** of `j` directly removes the contributions of edges the model already uses. Concretely, for each target `j` we form $S_j = \\mathrm{parents}_{G_\\mathrm{partial}}(j)$ and compute\n",
        "\n",
        "$$\\mathrm{pcorr}_b\\big(\\|z_i^{n-1}\\|,\\ \\|\\nabla_{z_j^n}\\mathcal{L}\\| \\mid \\{\\|z_k^{n-1}\\| : k \\in S_j\\}\\big)$$\n",
        "\n",
        "via Schur complement on the streamed `(sum_xx, sum_xy, sum_x2, sum_y2)` sufficient statistics — no extra forward/backward pass. The cost is one extra `(P, N, N)` accumulator and a small per-target solve at evaluation. p-values use Fisher-z with `df = n - 3 - |S_j|`. Kept-edge entries (`i ∈ S_j`) return NaN by construction, so `in_graph` rows drop out of the ranked output rather than competing with held-out candidates.\n",
        "\n",
        "### Validation\n",
        "\n",
        "This tutorial follows the synthetic recovery protocol (section 8.1): train on `G_partial ⊊ G*`, score all function-function pairs, and measure whether held-out edges rank above random non-edges (ROC AUC). **Within-target ranking** (`evaluate_target_ranking`) complements global AUC: for each target with a held-out parent, it asks whether the correct source is ranked first among all candidates for that target (MRR, top@k). A random-graph control (section 8.3) is a useful next step.\n",
        "\n",
        "### Normalization\n",
        "\n",
        "`MagnitudeEdgeInferer` uses **pre-norm** activations by default (`use_pre_norm=True`): magnitudes are computed after message passing (`lin_in`) but **before** layer / RMS / batch normalization. This preserves cross-sample magnitude variation even when the model is trained with `norm='layer'` or `norm='rms'`.\n",
        "\n",
        "Post-norm magnitudes (the old behavior, `use_pre_norm=False`) are still misleading for per-node norms — layer and RMS explicitly rescale each node's channels to unit scale, erasing the signal Tier 0 needs. Batch / groupbatch norms applied per channel still leave post-norm geometry somewhat usable, but pre-norm is the safer default for all norm types.\n",
        "\n",
        "### Next steps\n",
        "\n",
        "If AUC is at chance, try scoring from an earlier training checkpoint (before full convergence) or escalate to Tier 2 (learned scalar dual encoder). If AUC is strong, Tier 0 may be sufficient as a scalable screen before multivariate verification."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "17db5bce",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "id": "42157547",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "id": "6cec6346",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "id": "9fba2251",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "id": "a0e3aac8",
      "metadata": {},
      "source": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "gsnn-mds",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.12.11"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}