{
  "patent_number": "US 20220414425",
  "country": "US",
  "title": "How AI Automatically Builds Efficient Neural Networks for Limited Devices",
  "original_title": "Resource constrained neural network architecture search",
  "summary": "Google's patent describes a method for automatically designing efficient neural networks by optimizing their internal structure, called 'computational cells,' to meet specific limits on resources like memory or processing speed.",
  "what_it_does": "This patent details a computer-implemented method for automatically designing neural networks that are efficient for specific tasks and hardware. It starts by defining 'computational cells' (Claim 1), which are like small, reusable blueprints for parts of a neural network, composed of nodes (representing data) and edges (representing operations). For each operation, the system replaces it with a 'linear combination of candidate operations' (Claim 2), where each candidate operation has a 'mixing weight' controlled by 'computational cell hyper parameters.' The system then repeatedly adjusts these hyperparameters and weights to optimize how well the network performs (minimize 'validation loss function') while strictly adhering to 'computational resource constraints' (Claim 3, 9), such as limits on memory, processing power (float point operations), or how fast it can make predictions (inference speed). Finally, it uses these optimized settings to generate the complete neural network (Claim 1). For example, this could be used to create a highly efficient AI model for a smartwatch that needs to recognize speech without draining its small battery or exceeding its limited memory.",
  "what_it_does_not_cover": [
    "Does not cover manual design of neural network architectures by human engineers.",
    "Does not cover architecture search methods that do not use 'computational cells' defined as directed graphs of nodes and edges (Claim 1).",
    "Does not cover optimization methods that do not involve replacing operations with a 'linear combination of candidate operations' (Claim 2).",
    "Does not cover optimization without specific 'computational resource constraints' (Claim 1, 9) like memory, FLOPS, or inference speed.",
    "Does not cover methods that do not iteratively adjust 'computational cell hyper parameters and computational cell weights' during the optimization process (Claim 3)."
  ],
  "filed": "2022-08-19",
  "granted": null,
  "expires": "2042-08-19",
  "status": "active",
  "holder": "Google",
  "holder_url": "https://patentbrief.org/company/google",
  "inventors": [
    {
      "name": "Shengyang Dai",
      "url": "https://patentbrief.org/inventor/shengyang-dai"
    },
    {
      "name": "Joshua Foster Slocum",
      "url": "https://patentbrief.org/inventor/joshua-foster-slocum"
    },
    {
      "name": "Ming-Hsuan Yang",
      "url": "https://patentbrief.org/inventor/ming-hsuan-yang"
    },
    {
      "name": "Xiaojie Jin",
      "url": "https://patentbrief.org/inventor/xiaojie-jin"
    },
    {
      "name": "Jiang Wang",
      "url": "https://patentbrief.org/inventor/jiang-wang"
    }
  ],
  "times_cited": 2,
  "tags": [
    "software",
    "ai_ml",
    "consumer_electronics",
    "telecommunications",
    "semiconductors"
  ],
  "abstract": "Methods, and systems, including computer programs encoded on computer storage media for neural network architecture search. A method includes defining a neural network computational cell, the computational cell including a directed graph of nodes representing respective neural network latent representations and edges representing respective operations that transform a respective neural network latent representation; replacing each operation that transforms a respective neural network latent representation with a respective linear combination of candidate operations, where each candidate operation in a respective linear combination has a respective mixing weight that is parameterized by one or more computational cell hyper parameters; iteratively adjusting values of the computational cell hyper parameters and weights to optimize a validation loss function subject to computational resource constraints; and generating a neural network for performing a machine learning task using the defined computational cell and the adjusted values of the computational cell hyper parameters and weights.",
  "url": "https://patentbrief.org/patent/us/20220414425/resource-constrained-neural-network-architecture-search",
  "markdown_url": "https://patentbrief.org/patent/us/20220414425/resource-constrained-neural-network-architecture-search/md",
  "google_patents_url": "https://patents.google.com/patent/US20220414425",
  "relatedPatents": [
    {
      "patentNumber": "12632737",
      "countryCode": "US",
      "title": "How AI Designs Its Own Brains: Hierarchical Neural Architecture Search",
      "url": "https://patentbrief.org/patent/us/12632737/using-hierarchical-representations-for-neural-network-architecture-searching"
    },
    {
      "patentNumber": "20250371349",
      "countryCode": "US",
      "title": "How AI Learns to Run Faster on Specific Computer Chips",
      "url": "https://patentbrief.org/patent/us/20250371349/methods-and-apparatus-for-hardware-aware-machine-learning-model-training"
    },
    {
      "patentNumber": "20220383078",
      "countryCode": "US",
      "title": "Adapting AI Models to Fit Device Resources",
      "url": "https://patentbrief.org/patent/us/20220383078/data-processing-method-and-related-device"
    },
    {
      "patentNumber": "20250363357",
      "countryCode": "US",
      "title": "How to Update AI on Small Devices with Slow Internet",
      "url": "https://patentbrief.org/patent/us/20250363357/systems-and-methods-for-deploying-and-updating-neural-networks-at-the-edge-of-a-"
    },
    {
      "patentNumber": "20250292091",
      "countryCode": "US",
      "title": "How AI at the Edge Learns and Adapts Its Brain",
      "url": "https://patentbrief.org/patent/us/20250292091/systems-and-methods-for-dynamic-neural-network-enhancement-and-adaptive-edge-com"
    }
  ]
}