{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6d925cb3",
   "metadata": {},
   "source": [
    "# 15 — Le test adverse sous mesure consciente du rang, et l'estimation du biais de position\n",
    "\n",
    "Le [notebook 14](14_rang_et_contrefactuel.ipynb) a laissé deux dettes explicites.\n",
    "\n",
    "**La première.** Les quatre mesures comparées au [notebook 13](13_test_adverse_index.ipynb) ont\n",
    "été éprouvées sur des **compositions**, jamais sur des fils ordonnés. L'enterrement les concerne\n",
    "donc toutes, et aucune n'a été jugée avec lui.\n",
    "\n",
    "**La seconde.** Les estimateurs contrefactuels reposent sur un modèle de biais de position dont\n",
    "la sévérité $\\eta$ était **posée**, non mesurée.\n",
    "\n",
    "**Ce que ce notebook établit :**\n",
    "\n",
    "* sous un plancher aveugle au rang, **les quatre mesures se laissent contourner par\n",
    "  l'enterrement** — une plateforme certifiée à 0,70 n'expose que **0,36** ;\n",
    "* un plancher conscient du rang ferme l'échappatoire, au prix d'un coût d'engagement qui\n",
    "  **double** ;\n",
    "* $\\eta$ **s'estime** à partir des seules données enregistrées, à condition que la plateforme\n",
    "  ait varié ses classements — et le refus d'estimer, quand elle ne l'a pas fait, est aussi un\n",
    "  résultat ;\n",
    "* poser $\\eta$ de travers coûte jusqu'à **179 %** d'erreur, soit l'ordre de grandeur du biais\n",
    "  qu'on prétendait corriger. **Corriger ne suffit pas : il faut estimer la correction.**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "795e89ea",
   "metadata": {},
   "source": [
    "## 1. Les quatre mesures, cette fois sur des fils ordonnés\n",
    "\n",
    "Le fil compte $n$ positions à remplir depuis un catalogue de $k$ points de vue, soit $k^n$ fils\n",
    "possibles. Ils sont **tous énumérés** : l'optimum est exact, non le résultat d'une heuristique.\n",
    "\n",
    "Ce n'est pas un luxe. Il s'agit encore de résultats négatifs — des normes qui échouent — et un\n",
    "optimum manqué par un solveur y produirait exactement la même apparence qu'une norme qui tient."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2de8baa1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:18.879021Z",
     "iopub.status.busy": "2026-08-23T13:40:18.878797Z",
     "iopub.status.idle": "2026-08-23T13:40:19.420450Z",
     "shell.execute_reply": "2026-08-23T13:40:19.419945Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "65536 fils énumérés pour chaque norme\n",
      "engagement sans contrainte : 2.446\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from ide.gaming import (\n",
    "    canonical_positions,\n",
    "    gaussian_ild,\n",
    "    position_entropy,\n",
    "    rao_entropy,\n",
    "    target_divergence,\n",
    ")\n",
    "from ide.offpolicy import (\n",
    "    estimate_position_bias,\n",
    "    naive,\n",
    "    rank_propensities,\n",
    "    simulate_logged_feedback,\n",
    "    simulate_ranked_feedback,\n",
    "    snips,\n",
    "    value_under_policy,\n",
    ")\n",
    "from ide.plotting import PALETTE, save_figure, use_project_style\n",
    "from ide.ranking import all_rankings, optimal_ranking_under, ranking_engagement\n",
    "\n",
    "use_project_style()\n",
    "\n",
    "VIEWPOINTS, SLOTS = 4, 8\n",
    "CATALOGUE = canonical_positions(VIEWPOINTS)\n",
    "REFERENCE = float(np.ptp(CATALOGUE))\n",
    "BANDWIDTH = float(CATALOGUE[1] - CATALOGUE[0])\n",
    "\n",
    "# Un lecteur qui préfère nettement le premier point de vue : sans cette préférence, il n'y\n",
    "# aurait aucun intérêt à enterrer quoi que ce soit.\n",
    "RELEVANCE = np.array([0.90, 0.45, 0.25, 0.15])\n",
    "\n",
    "MEASURES = {\n",
    "    \"Rao (ILD)\": lambda w: rao_entropy(w, CATALOGUE, REFERENCE),\n",
    "    \"entropie de position\": lambda w: position_entropy(w, CATALOGUE, CATALOGUE),\n",
    "    \"Gaussian ILD\": lambda w: gaussian_ild(w, CATALOGUE, BANDWIDTH),\n",
    "    \"proximité à la cible\": lambda w: target_divergence(w, CATALOGUE, CATALOGUE),\n",
    "}\n",
    "\n",
    "unconstrained = ranking_engagement(np.zeros(SLOTS, dtype=int), RELEVANCE)\n",
    "print(f\"{VIEWPOINTS ** SLOTS} fils énumérés pour chaque norme\")\n",
    "print(f\"engagement sans contrainte : {unconstrained:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f19fda7b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:19.421288Z",
     "iopub.status.busy": "2026-08-23T13:40:19.421179Z",
     "iopub.status.idle": "2026-08-23T13:40:25.725436Z",
     "shell.execute_reply": "2026-08-23T13:40:25.724906Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "plancher imposé : 0.7\n",
      "\n",
      "mesure                        norme     coût   affiché   exposé   fil servi\n",
      "----------------------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rao (ILD)                   aveugle     8.2%     0.750    0.355   00000033\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rao (ILD)                consciente    18.9%     0.938    0.702   00033300\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "entropie de position        aveugle    10.7%     0.774    0.443   00000123\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "entropie de position     consciente    20.9%     0.953    0.702   00013122\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gaussian ILD                aveugle   plancher inatteignable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gaussian ILD             consciente   plancher inatteignable\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "proximité à la cible        aveugle     5.9%     0.750    0.628   00000012\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "proximité à la cible     consciente    10.6%     0.750    0.701   00100200\n"
     ]
    }
   ],
   "source": [
    "FLOOR = 0.70\n",
    "\n",
    "print(f\"plancher imposé : {FLOOR}\\n\")\n",
    "print(f\"{'mesure':22s} {'norme':>12s} {'coût':>8s} {'affiché':>9s} {'exposé':>8s}   fil servi\")\n",
    "print(\"-\" * 88)\n",
    "results = {}\n",
    "for name, measure in MEASURES.items():\n",
    "    for aware in (False, True):\n",
    "        best = optimal_ranking_under(measure, RELEVANCE, SLOTS, FLOOR, rank_aware=aware)\n",
    "        results[name, aware] = best\n",
    "        label = \"consciente\" if aware else \"aveugle\"\n",
    "        if best is None:\n",
    "            print(f\"{name:22s} {label:>12s}   plancher inatteignable\")\n",
    "            continue\n",
    "        cost = 100 * (1 - best.engagement / unconstrained)\n",
    "        print(f\"{name:22s} {label:>12s} {cost:7.1f}% {best.blind:9.3f} {best.aware:8.3f}\"\n",
    "              f\"   {''.join(map(str, best.assignment))}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4523eebd",
   "metadata": {},
   "source": [
    "### Lecture\n",
    "\n",
    "La colonne **affiché** est ce que la norme constate ; la colonne **exposé** est la diversité que\n",
    "le lecteur reçoit réellement, une fois l'attention de chaque rang prise en compte.\n",
    "\n",
    "Sous plancher **aveugle**, l'écart est béant. L'entropie de Rao certifie un fil à 0,750 dont la\n",
    "diversité exposée vaut **0,355** ; l'entropie de position certifie 0,774 pour 0,443. Le fil\n",
    "optimal est à chaque fois de la même forme — six contenus du point de vue préféré, puis les\n",
    "divergents relégués aux dernières positions.\n",
    "\n",
    "**Une plateforme certifiée à 0,70 n'expose donc que la moitié de ce qu'on lui compte.**\n",
    "\n",
    "Sous plancher **conscient du rang**, l'échappatoire est fermée : la diversité exposée atteint le\n",
    "plancher, et les contenus divergents remontent dans le fil. Le prix est réel — le coût\n",
    "d'engagement passe de 8,2 % à 18,9 % pour Rao, de 10,7 % à 20,9 % pour l'entropie de position.\n",
    "\n",
    "**La Gaussian ILD reste inatteignable** au-delà de 0,5 : sa borne dépend de $k$ et de la largeur\n",
    "de bande, ce que le notebook 13 avait déjà relevé comme l'empêchant de servir de seuil."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8a48b913",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:25.726326Z",
     "iopub.status.busy": "2026-08-23T13:40:25.726246Z",
     "iopub.status.idle": "2026-08-23T13:40:40.492333Z",
     "shell.execute_reply": "2026-08-23T13:40:40.491770Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Écart entre diversité affichée et diversité exposée, par plancher\n",
      "\n",
      " plancher           Rao (ILD)  entropie de positi        Gaussian ILD  proximité à la cib\n",
      "-----------------------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0.40               0.262               0.173               0.200               0.000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0.50               0.306               0.249               0.194               0.066\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0.60               0.350               0.263                   —               0.066\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0.70               0.395               0.331                   —               0.122\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0.80               0.397               0.319                   —               0.166\n",
      "\n",
      "L'écart n'est pas un artefact d'un plancher particulier : il croît avec l'exigence.\n",
      "Plus la norme aveugle demande de diversité, plus il devient rentable de l'enterrer.\n"
     ]
    }
   ],
   "source": [
    "print(\"Écart entre diversité affichée et diversité exposée, par plancher\\n\")\n",
    "print(f\"{'plancher':>9s}\" + \"\".join(f\"{n[:18]:>20s}\" for n in MEASURES))\n",
    "print(\"-\" * 89)\n",
    "scan = {}\n",
    "for floor in (0.4, 0.5, 0.6, 0.7, 0.8):\n",
    "    row = \"\"\n",
    "    for name, measure in MEASURES.items():\n",
    "        best = optimal_ranking_under(measure, RELEVANCE, SLOTS, floor, rank_aware=False)\n",
    "        scan[name, floor] = best\n",
    "        row += f\"{'—':>20s}\" if best is None else f\"{best.blind - best.aware:20.3f}\"\n",
    "    print(f\"{floor:9.2f}{row}\")\n",
    "\n",
    "print(\"\\nL'écart n'est pas un artefact d'un plancher particulier : il croît avec l'exigence.\")\n",
    "print(\"Plus la norme aveugle demande de diversité, plus il devient rentable de l'enterrer.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5681a61c",
   "metadata": {},
   "source": [
    "### Une mesure se compare à elle-même\n",
    "\n",
    "Le [test adverse](../docs/gaming.md) avait dû renoncer à seuiller l'écart entre deux indices\n",
    "*différents* — IDE et Rao ne sont pas sur la même échelle, et un fil honnête en affichait déjà\n",
    "0,36. L'écart mesuré ici est d'une autre nature : c'est **la même mesure appliquée deux fois au\n",
    "même fil**, une fois à l'aveugle du rang et une fois en le prenant en compte.\n",
    "\n",
    "Il vaut zéro pour un fil dont l'ordre ne concentre pas l'attention, et il est directement\n",
    "interprétable — ce qui en fait, cette fois, une grandeur seuillable."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a1b6f60",
   "metadata": {},
   "source": [
    "## 2. Estimer $\\eta$ plutôt que le poser\n",
    "\n",
    "Le modèle à biais de position pose $P(\\text{clic}) = R^{-\\eta}\\,g(i)$, donc\n",
    "\n",
    "$$\\log \\mathrm{CTR}(i, R) = \\log g(i) - \\eta \\log R$$\n",
    "\n",
    "La pertinence $g(i)$ y est un **effet fixe de contenu** : on ne cherche pas à l'estimer, on\n",
    "l'élimine en centrant à l'intérieur de chaque contenu. Ce qui subsiste est la seule variation\n",
    "qui identifie $\\eta$ — celle d'un **même contenu vu à des rangs différents**.\n",
    "\n",
    "C'est la forme la plus simple de la récolte d'interventions : elle n'exige aucune expérience,\n",
    "seulement que la plateforme n'ait pas toujours classé les mêmes contenus aux mêmes places."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2c299636",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:40.493210Z",
     "iopub.status.busy": "2026-08-23T13:40:40.493125Z",
     "iopub.status.idle": "2026-08-23T13:40:41.671576Z",
     "shell.execute_reply": "2026-08-23T13:40:41.671026Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  η vrai   η estimé  erreur type  contenus variables\n",
      "----------------------------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    0.40      0.409       0.0045                  12\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    0.70      0.693       0.0064                  12\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    1.00      1.005       0.0085                  12\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    1.30      1.315       0.0121                  12\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    1.60      1.649       0.0192                  12\n"
     ]
    }
   ],
   "source": [
    "rng = np.random.default_rng(5)\n",
    "catalogue_relevance = rng.uniform(0.15, 0.9, 12)\n",
    "\n",
    "print(f\"{'η vrai':>8s} {'η estimé':>10s} {'erreur type':>12s} {'contenus variables':>19s}\")\n",
    "print(\"-\" * 52)\n",
    "for true_severity in (0.4, 0.7, 1.0, 1.3, 1.6):\n",
    "    items, ranks, clicks = simulate_ranked_feedback(\n",
    "        catalogue_relevance, 40_000, true_severity, exploration=0.5, rng=rng\n",
    "    )\n",
    "    estimate = estimate_position_bias(items, ranks, clicks)\n",
    "    print(f\"{true_severity:8.2f} {estimate.severity:10.3f} {estimate.standard_error:12.4f}\"\n",
    "          f\" {estimate.items_with_variation:19d}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ddb030a3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:41.672358Z",
     "iopub.status.busy": "2026-08-23T13:40:41.672287Z",
     "iopub.status.idle": "2026-08-23T13:40:43.049880Z",
     "shell.execute_reply": "2026-08-23T13:40:43.049349Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "L'exploration de la plateforme est la condition d'identifiabilité\n",
      "\n",
      " exploration   η estimé  erreur type  variables  identifiable\n",
      "------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        0.00        n/a          n/a          0  NON\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        0.02      1.595       0.2468         10  oui\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        0.05      0.800       0.0923         12  oui\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        0.15      1.031       0.0364         12  oui\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        0.50      1.004       0.0112         12  oui\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        1.50      0.990       0.0083         12  oui\n",
      "\n",
      "À exploration nulle, aucun contenu ne change de rang : le paramètre n'est pas dans\n",
      "les données, et la fonction refuse de renvoyer un chiffre plutôt que d'en inventer un.\n",
      "À exploration faible, elle en renvoie un — mais l'erreur type dit qu'il ne vaut rien.\n"
     ]
    }
   ],
   "source": [
    "print(\"L'exploration de la plateforme est la condition d'identifiabilité\\n\")\n",
    "print(f\"{'exploration':>12s} {'η estimé':>10s} {'erreur type':>12s} {'variables':>10s}  identifiable\")\n",
    "print(\"-\" * 66)\n",
    "identifiability = []\n",
    "for exploration in (0.0, 0.02, 0.05, 0.15, 0.5, 1.5):\n",
    "    items, ranks, clicks = simulate_ranked_feedback(\n",
    "        catalogue_relevance, 40_000, 1.0, exploration=exploration, rng=rng\n",
    "    )\n",
    "    estimate = estimate_position_bias(items, ranks, clicks)\n",
    "    identifiability.append((exploration, estimate))\n",
    "    severity = f\"{estimate.severity:10.3f}\" if estimate.identifiable else f\"{'n/a':>10s}\"\n",
    "    error = f\"{estimate.standard_error:12.4f}\" if estimate.identifiable else f\"{'n/a':>12s}\"\n",
    "    print(f\"{exploration:12.2f} {severity} {error} {estimate.items_with_variation:10d}\"\n",
    "          f\"  {'oui' if estimate.identifiable else 'NON'}\")\n",
    "\n",
    "print(\"\\nÀ exploration nulle, aucun contenu ne change de rang : le paramètre n'est pas dans\")\n",
    "print(\"les données, et la fonction refuse de renvoyer un chiffre plutôt que d'en inventer un.\")\n",
    "print(\"À exploration faible, elle en renvoie un — mais l'erreur type dit qu'il ne vaut rien.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2337da22",
   "metadata": {},
   "source": [
    "## 3. Pourquoi il fallait l'estimer\n",
    "\n",
    "La question qui relie les deux moitiés de ce notebook : **que coûte un $\\eta$ posé de travers ?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "145cf2f2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:43.050714Z",
     "iopub.status.busy": "2026-08-23T13:40:43.050637Z",
     "iopub.status.idle": "2026-08-23T13:40:43.071660Z",
     "shell.execute_reply": "2026-08-23T13:40:43.071249Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "coût réel du filtre de diversité : 6.6 %\n",
      "\n",
      "  η supposé  coût estimé    erreur\n",
      "------------------------------------\n",
      "        0.5         2.7 %    -59.0 %\n",
      "        0.8         4.9 %    -25.6 %\n",
      "        1.0         6.6 %      0.6 %\n",
      "        1.2         8.6 %     30.1 %\n",
      "        1.5        11.9 %     80.3 %\n",
      "        2.0        18.4 %    178.8 %\n"
     ]
    }
   ],
   "source": [
    "ITEMS, IMPRESSIONS, TRUE_SEVERITY = 20, 400_000, 1.0\n",
    "rng = np.random.default_rng(11)\n",
    "\n",
    "relevance = rng.uniform(0.05, 0.95, ITEMS)\n",
    "diversity = rng.uniform(0.0, 1.0, ITEMS)\n",
    "logged_ranks = np.argsort(np.argsort(-relevance)) + 1\n",
    "target_ranks = np.argsort(np.argsort(-(0.4 * relevance + 0.6 * diversity))) + 1\n",
    "\n",
    "logged = rank_propensities(logged_ranks, TRUE_SEVERITY)\n",
    "target = rank_propensities(target_ranks, TRUE_SEVERITY)\n",
    "examined, clicks = simulate_logged_feedback(relevance, logged, IMPRESSIONS, rng)\n",
    "true_cost = 1 - value_under_policy(relevance, target) / value_under_policy(relevance, logged)\n",
    "\n",
    "\n",
    "def cost_assuming(severity):\n",
    "    assumed_logged = rank_propensities(logged_ranks, severity)\n",
    "    assumed_target = rank_propensities(target_ranks, severity)\n",
    "    return 1 - snips(clicks, assumed_target[examined], assumed_logged[examined]) / naive(clicks)\n",
    "\n",
    "\n",
    "print(f\"coût réel du filtre de diversité : {100 * true_cost:.1f} %\\n\")\n",
    "print(f\"{'η supposé':>11s} {'coût estimé':>12s} {'erreur':>9s}\")\n",
    "print(\"-\" * 36)\n",
    "sensitivity = []\n",
    "for severity in (0.5, 0.8, 1.0, 1.2, 1.5, 2.0):\n",
    "    estimated = cost_assuming(severity)\n",
    "    sensitivity.append((severity, estimated))\n",
    "    print(f\"{severity:11.1f} {100 * estimated:11.1f} % {100 * (estimated - true_cost) / true_cost:8.1f} %\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "27a610d3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:43.072453Z",
     "iopub.status.busy": "2026-08-23T13:40:43.072383Z",
     "iopub.status.idle": "2026-08-23T13:40:43.511981Z",
     "shell.execute_reply": "2026-08-23T13:40:43.511471Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "η estimé sur les données enregistrées : 1.013 ± 0.019\n",
      "\n",
      "  coût estimé sur cet intervalle : 6.6 % à 6.9 %\n",
      "  coût réel                      : 6.6 %\n",
      "\n",
      "L'incertitude sur η devient une bande de quelques dixièmes de point sur le résultat,\n",
      "là où un η posé au jugé pouvait le tripler.\n"
     ]
    }
   ],
   "source": [
    "items, ranks, click_flags = simulate_ranked_feedback(\n",
    "    relevance, 40_000, TRUE_SEVERITY, exploration=0.5, rng=rng\n",
    ")\n",
    "measured = estimate_position_bias(items, ranks, click_flags)\n",
    "low = measured.severity - 2 * measured.standard_error\n",
    "high = measured.severity + 2 * measured.standard_error\n",
    "band = [cost_assuming(value) for value in (low, measured.severity, high)]\n",
    "\n",
    "print(f\"η estimé sur les données enregistrées : {measured.severity:.3f}\"\n",
    "      f\" ± {2 * measured.standard_error:.3f}\\n\")\n",
    "print(f\"  coût estimé sur cet intervalle : {100 * min(band):.1f} % à {100 * max(band):.1f} %\")\n",
    "print(f\"  coût réel                      : {100 * true_cost:.1f} %\")\n",
    "print(\"\\nL'incertitude sur η devient une bande de quelques dixièmes de point sur le résultat,\")\n",
    "print(\"là où un η posé au jugé pouvait le tripler.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ee36e955",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:40:43.512803Z",
     "iopub.status.busy": "2026-08-23T13:40:43.512731Z",
     "iopub.status.idle": "2026-08-23T13:40:43.941274Z",
     "shell.execute_reply": "2026-08-23T13:40:43.940803Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1380x936 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure, axes = plt.subplots(2, 2, figsize=(11.5, 7.8))\n",
    "exposure, spread, recovery, sensitivity_panel = axes.ravel()\n",
    "\n",
    "# (a) Affiché contre exposé, sous plancher aveugle puis conscient.\n",
    "names = [n for n in MEASURES if results[n, False] is not None]\n",
    "positions = np.arange(len(names))\n",
    "displayed = [results[n, False].blind for n in names]\n",
    "received = [results[n, False].aware for n in names]\n",
    "exposure.bar(positions - 0.19, displayed, 0.36, color=PALETTE[\"neutral\"],\n",
    "             label=\"diversité affichée\")\n",
    "exposure.bar(positions + 0.19, received, 0.36, color=PALETTE[\"disorder\"],\n",
    "             label=\"diversité réellement exposée\")\n",
    "exposure.axhline(FLOOR, color=PALETTE[\"order\"], linestyle=\"--\", linewidth=1.5)\n",
    "exposure.text(len(names) - 0.55, FLOOR + 0.02, f\"plancher {FLOOR:.2f}\", fontsize=8,\n",
    "              color=PALETTE[\"order\"], ha=\"right\")\n",
    "exposure.set_xticks(positions)\n",
    "exposure.set_xticklabels([n.replace(\" \", \"\\n\") for n in names], fontsize=7.5)\n",
    "exposure.set_ylabel(\"valeur de la mesure\")\n",
    "exposure.set_ylim(0, 1.05)\n",
    "exposure.set_title(\"Sous plancher aveugle, l'exposé est bien moindre\", fontsize=10)\n",
    "exposure.legend(fontsize=7.5, loc=\"upper right\")\n",
    "\n",
    "# (b) L'écart croît avec l'exigence de la norme.\n",
    "floors = [0.4, 0.5, 0.6, 0.7, 0.8]\n",
    "palette = [PALETTE[\"disorder\"], PALETTE[\"remedy\"], PALETTE[\"neutral\"], PALETTE[\"order\"]]\n",
    "for (name, _), colour in zip(MEASURES.items(), palette, strict=True):\n",
    "    curve = [scan[name, floor].blind - scan[name, floor].aware\n",
    "             if scan[name, floor] is not None else np.nan for floor in floors]\n",
    "    spread.plot(floors, curve, marker=\"o\", markersize=4, linewidth=1.7, color=colour, label=name)\n",
    "spread.set_xlabel(\"plancher aveugle imposé\")\n",
    "spread.set_ylabel(\"écart affiché − exposé\")\n",
    "spread.set_title(\"Plus la norme exige, plus enterrer rapporte\", fontsize=10)\n",
    "spread.legend(fontsize=7.5)\n",
    "\n",
    "# (c) La sévérité retrouvée, et le seuil d'identifiabilité.\n",
    "explorations = [value for value, estimate in identifiability if estimate.identifiable]\n",
    "estimates = [estimate.severity for _, estimate in identifiability if estimate.identifiable]\n",
    "errors = [2 * estimate.standard_error for _, estimate in identifiability if estimate.identifiable]\n",
    "recovery.errorbar(explorations, estimates, yerr=errors, marker=\"o\", markersize=5,\n",
    "                  linewidth=1.6, capsize=3, color=PALETTE[\"remedy\"])\n",
    "recovery.axhline(1.0, color=PALETTE[\"order\"], linestyle=\"--\", linewidth=1.5)\n",
    "recovery.text(0.55, 1.02, \"η vrai\", fontsize=8, color=PALETTE[\"order\"])\n",
    "recovery.axvline(0.0, color=PALETTE[\"disorder\"], linewidth=2.5)\n",
    "recovery.text(0.024, 0.66, \"plateforme déterministe :\\nη non identifiable\", fontsize=8,\n",
    "              color=PALETTE[\"disorder\"], va=\"bottom\")\n",
    "recovery.set_xscale(\"symlog\", linthresh=0.02)\n",
    "recovery.set_xlim(-0.004, 2.2)\n",
    "recovery.set_xlabel(\"exploration du classement de la plateforme\")\n",
    "recovery.set_ylabel(\"η estimé\")\n",
    "recovery.set_title(\"η s'estime, si la plateforme a varié ses rangs\", fontsize=10)\n",
    "\n",
    "# (d) Ce que coûte un η posé au jugé.\n",
    "assumed = [value for value, _ in sensitivity]\n",
    "estimated = [100 * value for _, value in sensitivity]\n",
    "sensitivity_panel.plot(assumed, estimated, marker=\"o\", markersize=5, linewidth=1.8,\n",
    "                       color=PALETTE[\"field\"], label=\"coût estimé\")\n",
    "sensitivity_panel.axhline(100 * true_cost, color=PALETTE[\"order\"], linestyle=\"--\", linewidth=1.5)\n",
    "sensitivity_panel.text(0.52, 100 * true_cost + 0.5, \"coût réel\", fontsize=8,\n",
    "                       color=PALETTE[\"order\"])\n",
    "sensitivity_panel.axvspan(low, high, color=PALETTE[\"remedy\"], alpha=0.25)\n",
    "sensitivity_panel.text((low + high) / 2, 17.0, \"η estimé\\n± 2 erreurs types\", fontsize=7.5,\n",
    "                       color=PALETTE[\"remedy\"], ha=\"center\")\n",
    "sensitivity_panel.set_xlabel(\"η supposé par l'estimateur\")\n",
    "sensitivity_panel.set_ylabel(\"coût estimé du filtre  [%]\")\n",
    "sensitivity_panel.set_title(\"Corriger avec le mauvais η corrige mal\", fontsize=10)\n",
    "sensitivity_panel.legend(fontsize=8, loc=\"upper left\")\n",
    "\n",
    "figure.suptitle(\"Rang adverse et sévérité : ce qu'une norme aveugle laisse passer\", fontsize=12)\n",
    "figure.tight_layout(rect=(0, 0, 1, 0.96))\n",
    "save_figure(figure, \"fig15_rang_adverse\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67ea57e7",
   "metadata": {},
   "source": [
    "## 4. Ce que le notebook établit\n",
    "\n",
    "**Les quatre mesures se laissent contourner par l'enterrement.** Jugées sur des fils ordonnés\n",
    "plutôt que sur des compositions, toutes certifient une diversité que le lecteur ne reçoit pas.\n",
    "À plancher 0,70, l'entropie de Rao certifie 0,750 pour une diversité exposée de **0,355**. Le\n",
    "test adverse initial ne les avait pas éprouvées contre cet adversaire — c'est fait.\n",
    "\n",
    "**Un plancher conscient du rang le ferme, et il coûte.** Le prix d'engagement double : de 8,2 %\n",
    "à 18,9 % pour l'entropie de Rao, de 10,7 % à 20,9 % pour l'entropie de position. Une norme qui\n",
    "ne coûterait pas davantage ne fermerait rien.\n",
    "\n",
    "**L'écart affiché − exposé est cette fois seuillable.** Contrairement à l'écart entre deux\n",
    "indices différents, que le test adverse avait dû renoncer à seuiller, celui-ci compare la même\n",
    "mesure à elle-même. Il vaut zéro pour un fil qui ne relègue pas, et croît avec l'exigence de la\n",
    "norme aveugle — plus elle demande, plus enterrer rapporte.\n",
    "\n",
    "**La sévérité du biais de position s'estime**, à ±0,02 sur 40 000 impressions, par une\n",
    "régression à effets fixes de contenu. Mais seulement si la plateforme **a varié ses\n",
    "classements** : à politique déterministe, le paramètre n'est pas dans les données et\n",
    "l'estimateur refuse de renvoyer un chiffre.\n",
    "\n",
    "**Et il fallait l'estimer.** Poser $\\eta = 0{,}5$ au lieu de 1,0 donne une erreur de −59 % ;\n",
    "poser 2,0 donne +179 %. C'est l'ordre de grandeur du biais qu'on prétendait corriger.\n",
    "\n",
    "> **Corriger ne suffit pas : il faut estimer le paramètre de la correction, et publier son\n",
    "> incertitude.** Sur l'intervalle estimé, le coût tient entre 6,6 % et 6,9 % — contre une\n",
    "> fourchette de 2,7 % à 18,4 % si l'on pose $\\eta$ au jugé.\n",
    "\n",
    "### Les hypothèses qui restent\n",
    "\n",
    "La forme $e(R) = R^{-\\eta}$ est **posée**, et seule sa sévérité est estimée. Un modèle\n",
    "d'examen plus riche — dépendant du contenu, ou de ce que le lecteur a déjà vu — donnerait des\n",
    "propensions différentes.\n",
    "\n",
    "L'énumération exhaustive borne par ailleurs la taille des fils étudiés : huit positions sur\n",
    "quatre points de vue. Rien n'assure que le comportement observé se transporte à un fil de\n",
    "cinquante items sur cinquante points de vue, et l'affirmer demanderait une optimisation dont\n",
    "l'exactitude ne serait plus garantie."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d4e1eea",
   "metadata": {},
   "source": [
    "## Pistes ouvertes\n",
    "\n",
    "1. **Vérifier que l'enterrement se transporte à grande échelle.** L'énumération exhaustive\n",
    "   s'arrête à quelques dizaines de milliers de fils ; au-delà, il faudrait une optimisation\n",
    "   approchée, dont il faudrait alors établir qu'elle n'invente pas le résultat.\n",
    "2. **Enrichir le modèle d'examen.** La sévérité est estimée sous l'hypothèse que l'attention ne\n",
    "   dépend que du rang. Les modèles à confiance ou à cascade la font dépendre aussi de ce que le\n",
    "   lecteur a déjà consulté.\n",
    "3. **Mesurer l'exploration réelle d'un jeu de données public** avant d'en tirer quoi que ce\n",
    "   soit : c'est elle qui décide si $\\eta$ y est identifiable, et donc si l'évaluation\n",
    "   contrefactuelle y est possible.\n",
    "4. **Comparer à des lignes de base réglées** — MMR, réordonnancement aléatoire, popularité —\n",
    "   qui reste la dette du programme d'évaluation."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.12.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}