{
  "docId": "019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "docSlug": "bf350dd574c19997",
  "documentTitle": "2021 Air Street Capital The State of AI Report 2021",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 108,
  "pageCount": 188,
  "prevPage": 107,
  "nextPage": 109,
  "slideType": "industry_trends",
  "function": "establish_context",
  "density": "overcrowded",
  "nDataPoints": 24,
  "notes": "The slide highlights the importance of robust benchmarks for real-world ML applications.",
  "elementsJson": [
    "bullet_list",
    "data_table",
    "infographic"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5c-fd5012384ae3/108",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3#slide-108",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Two new datasets to deal with distribution shifts: WILDS and Shifts.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a20a-ed30dc660a27",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.28,
        "x": 0.68,
        "y": 0.45
      },
      "kind": "image",
      "text": "WILDS dataset visual examples",
      "attrs": null,
      "subkind": "infographic",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "5490c74b-34a7-4efd-8e20-2566c80d36bc",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.12,
        "w": 0.6,
        "x": 0.03,
        "y": 0.48
      },
      "kind": "list",
      "text": "A team from several American and Japanese universities and companies have built WILDS, a benchmark of 10 datasets of distributional shifts in tumor identification, wildlife monitoring, satellite imaging, and more.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "3c2d08ec-fe2c-46e8-b111-7aa862925dc8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.6,
        "x": 0.03,
        "y": 0.6
      },
      "kind": "list",
      "text": "Shifts, developed by the Russian Yandex, is more industry-focused, and includes 3 tasks: weather prediction, machine translation and vehicle motion prediction.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "a33ebb68-7483-4c10-a6c8-cbf76c64a20b",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.6,
        "x": 0.03,
        "y": 0.4
      },
      "kind": "list",
      "text": "As ML is increasingly used in real-world applications, the need for a solid understanding of distributional shifts becomes paramount. This begins with designing challenging benchmarks.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d8489064-bf67-49e1-a077-e15526962028",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.12,
        "w": 0.6,
        "x": 0.03,
        "y": 0.28
      },
      "kind": "list",
      "text": "A distribution shift happens when data at test/deployment time is different from the training data. In production, this often happens in the form of concept drifts, where the test data gradually changes over time.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "ee58cd7b-61c3-49c0-8cde-ffc485633b69",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Number of samples: 3,129,592",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a20a-f27070e1a09a",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.15,
        "w": 0.4,
        "x": 0.27,
        "y": 0.78
      },
      "kind": "table",
      "text": "Table 1: Number of samples in the canonical partitioning of Weather Prediction dataset.",
      "attrs": null,
      "subkind": "data",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "90a4d276-db2b-4a6d-8675-c08754902c0e",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.04,
        "w": 0.55,
        "x": 0.03,
        "y": 0.22
      },
      "kind": "title",
      "text": "Two new datasets to deal with distribution shifts: WILDS and Shifts.",
      "attrs": null,
      "subkind": "action-title",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "e30cda89-e42f-4faa-9790-2290c4e0ec37",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.5,
        "x": 0.03,
        "y": 0.14
      },
      "kind": "title",
      "text": "Machine Learning in production: distribution shifts",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "be929087-5569-404c-8e7a-a87499da344b",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 152,
      "from": 5,
      "beatId": "019dd95a-0682-776c-8e35-0affbadec38e",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Definitions p.5-6, Exec Summary p.7, then Sections 1-3 catalog evidence.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 152,
      "from": 10,
      "beatId": "019dd95a-0682-776c-8e35-1b3ee3a1c012",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Three sections accumulating evidence of accelerating progress.",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 111,
      "from": 106,
      "name": "Iceberg",
      "slug": "10-iceberg",
      "bestFor": "Consulting, complex problem solving, organizational change",
      "matchId": "019dd95a-07fe-70ce-8d3d-77cba6708c34",
      "evidence": "Data-centric pivot, benchmarking, distribution shifts, underspecification, bad data, data-driven AI.",
      "position": 14,
      "objective": "Diagnose ML-in-production challenges",
      "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause",
      "confidence": 78,
      "description": "Reveal that the visible problem is merely a symptom of a deeper root cause"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}