{
  "docId": "019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "docSlug": "cc8183ec02431b7a",
  "documentTitle": "2022 Air Street Capital The State of AI Report 2022",
  "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": 38,
  "pageCount": 114,
  "prevPage": 37,
  "nextPage": 39,
  "slideType": "industry_trends",
  "function": "analyze_data",
  "density": "dense",
  "nDataPoints": 8,
  "notes": "The chart compares model performance (ImageNet accuracy) vs computational cost (GFLOPs) for various architectures.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "bubble_chart"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc/38",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc#slide-38",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Not so fast: work from Meta and UC Berkeley argues that modernizing ConvNets gives them an edge over ViTs.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a217-55ec819b17c2",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.4,
        "w": 0.35,
        "x": 0.62,
        "y": 0.45
      },
      "kind": "chart",
      "text": "Bubble chart showing ImageNet-1K accuracy vs GFLOPs for various models.",
      "attrs": null,
      "subkind": "bubble",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "973529cc-74da-4c47-a5cb-c330402fb733",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.65,
        "w": 0.55,
        "x": 0.05,
        "y": 0.22
      },
      "kind": "list",
      "text": "The introduction of Vision Transformers (ViT) and other image transformers last year as SOTA models on imaging benchmarks announced the dawn of ConvNets. Not so fast: work from Meta and UC Berkeley argues that modernizing ConvNets gives them an edge over ViTs.\nThe researchers introduce ConvNeXt, a ResNet which is augmented with the recent design choices introduced in hierarchical vision Transformers like Swin, but doesn't use attention layers.\nConvNeXt is both competitive with Swin Transformer and ViT on ImageNet-1K and ImageNet-22K and benefits from scale like them.\nTransformers quickly replaced recurrent neural networks in language modeling, but we don't expect a similar abrupt drop-off in ConvNets usage, especially in smaller scale ML use-cases.\nMeanwhile, our 2021 prediction of small transformers + CNN hybrid models manifested in MaxViT from Google with 475M parameters almost matching (89.53%) CoAtNet-7's performance (90.88%) on ImageNet top-1 accuracy.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "dbcccf44-2275-4851-9992-82a7fd2552fd",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "ImageNet-1K Accuracy: 89.53%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a217-5a919edc6b7b",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.9,
        "x": 0.05,
        "y": 0.13
      },
      "kind": "title",
      "text": "2021 Prediction: in vision, convolutional networks want a fair fight with transformers...",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "9283a7b7-52dc-423e-b509-5adfb3db559b",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Chart Selection Guide",
      "slug": "chart-selection-guide",
      "agent": "Designer",
      "layer": "slide",
      "matchId": "fe0edb40-62e4-473a-a921-6ee94c417c99",
      "evidence": "The slide features a bubble chart, comparing ImageNet-1K accuracy vs GFLOPs for various models.",
      "confidence": 0.7
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 80,
      "from": 5,
      "beatId": "019dd95a-0682-776c-8e35-2557e3799e96",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Definitions, exec summary, Research and Industry sections inventory the state of AI.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 93,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-35abf646aa8b",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Research, Industry, Politics sections accumulate signals.",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 43,
      "from": 37,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-afe811e71ad6",
      "evidence": "SayCan, ConvNets vs ViT, vision-language unification, Gato, IRIS, NeRFs as parallel evidence.",
      "position": 6,
      "objective": "Show transformers becoming the universal cross-modality architecture",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 78,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}