Nobel Prize in Physics awarded for training artificial neural networks (ANNs) using physics | Current Affairs | Vision IAS
News Today Logo

Nobel Prize in Physics awarded for training artificial neural networks (ANNs) using physics

Posted 09 Oct 2024

2 min read

It has been awarded jointly to John Hopfield & Geoffrey Hinton for constructing methods that helped lay foundation for machine learning (A type of AI) using ANNs.

What are ANNs?

  • ANNs are a subset of Machine Learning algorithms designed to model workings of human brain. 
  • ANNs consist of interconnected nodes, or artificial neurons, that process information similarly to how neurons function in the human brain.

Discoveries

  • John Hopfield: He invented a type of neural network (Hopfield Network) which is designed to store and recall patterns, similar to how memory works. 
    • Hopfield network utilizes physics that describes a material’s characteristics due to its atomic spin.
      • Atomic spin is magnetic moment of an atom that is caused by spins of particles that make up atoms.
  • Geoffrey Hinton: He invented a method (Boltzman Machine) that can autonomously find properties in data e.g. identifying specific elements in pictures.
    • Boltzmann machine learns by using examples that it may see while it works. It can sort images or create new patterns similar to what it learned.
      • This network uses methods from statistical physics.

Role of ANNs in AI

  • Deep Learning: ANNs are foundation of Deep Learning, a branch of Machine Learning that deals with large datasets and complex models. 
  • Use in AI: AI systems used for image recognition, natural language processing, and autonomous systems rely on ANNs to make decisions without human intervention.
  • Learning from Data: For example, ANN trained on thousands of medical images can eventually detect tumors in new images with high accuracy.
  • Tags :
  • Artificial Neural Networks
  • Hopfield Network
  • Boltzman Machine
Watch News Today