Samsung electronics co., ltd. (20240160892). METHOD AND SYSTEM FOR SELECTING AN ARTIFICIAL INTELLIGENCE (AI) MODEL IN NEURAL ARCHITECTURE SEARCH (NAS) simplified abstract

From WikiPatents
Revision as of 02:53, 23 May 2024 by Wikipatents (talk | contribs) (Creating a new page)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

METHOD AND SYSTEM FOR SELECTING AN ARTIFICIAL INTELLIGENCE (AI) MODEL IN NEURAL ARCHITECTURE SEARCH (NAS)

Organization Name

samsung electronics co., ltd.

Inventor(s)

Prateek Keserwani of Bengaluru (IN)

Srinivas Soumitri Miriyala of Bengaluru (IN)

Vikram Nelvoy Rajendiran of Bengaluru (IN)

Pradeep Nelahonne Shivamurthappa of Bangaluru (IN)

METHOD AND SYSTEM FOR SELECTING AN ARTIFICIAL INTELLIGENCE (AI) MODEL IN NEURAL ARCHITECTURE SEARCH (NAS) - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240160892 titled 'METHOD AND SYSTEM FOR SELECTING AN ARTIFICIAL INTELLIGENCE (AI) MODEL IN NEURAL ARCHITECTURE SEARCH (NAS)

Simplified Explanation

The abstract describes a method for selecting an artificial intelligence (AI) model in neural architecture search based on the scale of the receptive field of neural network layers. Here is a simplified explanation of the abstract:

  • Measuring the receptive field scale for neural network layers in candidate AI models
  • Calculating scores for groups of layers based on receptive field scale
  • Determining a final score for each candidate AI model
  • Selecting the AI model with the highest final score for training and deployment
      1. Potential Applications

This technology could be applied in various fields such as computer vision, natural language processing, and autonomous systems where AI models need to be optimized for specific tasks.

      1. Problems Solved

This method helps in efficiently selecting the most suitable AI model for a given task by considering the receptive field scale of neural network layers, leading to improved performance and accuracy.

      1. Benefits

- Enhanced performance of AI models - Streamlined process for selecting optimal AI models - Increased efficiency in training and deployment of AI systems

      1. Potential Commercial Applications

The technology could be utilized in industries such as healthcare, finance, and e-commerce for developing AI solutions tailored to specific needs, leading to better outcomes and competitive advantages.

      1. Possible Prior Art

Prior research in neural architecture search and AI model selection may exist, but this specific method focusing on receptive field scale as a criterion for model selection appears to be a novel approach.

        1. Unanswered Questions
        1. How does this method compare to existing techniques for AI model selection?

This article does not provide a direct comparison with other methods for selecting AI models based on receptive field scale. Further research or experimentation may be needed to evaluate its effectiveness in comparison to existing techniques.

        1. What are the potential limitations or challenges of implementing this method in real-world AI applications?

The abstract does not address any potential drawbacks or challenges that may arise when implementing this method in practical AI systems. It would be important to investigate factors such as computational complexity, scalability, and generalizability to different tasks.


Original Abstract Submitted

a method for selecting an artificial intelligence (ai) model in neural architecture search, includes: measuring a scale of receptive field for a plurality of neural network layers corresponding to each of a plurality of candidate ai models; determining a first score for a first group of neural network layers among the plurality of neural network layers based on the scale of the receptive field for the first group of neural network layers, the scale of the receptive field for each of the first group of neural network layers being smaller than a size of an object; determining a second score for a second group of neural network layers among the plurality of neural network layers based on the scale of the receptive field for the second group of neural network layers, the scale of the receptive field for each of the second group of neural network layers being greater than the size of the object; determining a third score for each of the plurality of candidate ai models as a function of the first score and the second score; and selecting, based on the third score, a candidate ai model among the plurality of candidate ai models for training and deployment, the candidate ai model having a highest third score among the third scores of the plurality of candidate ai models.