The Nobel Prize in Physics was awarded to two scientists, Geoffrey Hinton and John Hopfield, for their groundbreaking work that laid the foundations of artificial intelligence. The prize jury highlighted that their discoveries enabled the development of machine learning using artificial neural networksa technology integrated into widely used tools like ChatGPT.

Understanding Neural Networks and Machine Learning

But what is a neural network and how does machine learning work? Mark van der Wilk, a machine learning expert at the University of Oxford, explains that an artificial neural network is a mathematical construct “loosely inspired” by the functioning of the human brain.

Our brain is home to a complex network of neurons that react to stimuli coming from our environment. This learning process is based on strengthening or weakening the connections between these neurons. Unlike traditional computing, which follows precise instructions, artificial neural networks rely on training data to learn and adapt.

Contributions by John Hopfield

Before machines could actually learn, another essential characteristic was needed: memory. In the 1980s, John Hopfield developed the concept of Hopfield networkor “associative memory”.

When this type of network receives partial information, it is able to fill in the “empty spaces” based on previous models. This ability to find information from incomplete data constitutes a significant advance for artificial intelligence.

The Boltzmann machine by Geoffrey Hinton

Geoffrey Hinton, often called the “godfather of AI”, made his own major contribution with the Boltzmann machine in 1985. This model introduces an element of chance which plays a crucial role in the operation of current image generators. His research also demonstrated that adding additional layers to a neural network increases the complexity of learned behaviors.

Applications of Artificial Intelligence

Despite their potential, Hinton and Hopfield’s ideas did not immediately spark interest. In the 1990s, machine learning was limited by the capabilities of computers at the time, requiring powerful tools capable of processing enormous amounts of data. It was only from the 2010s that major breakthroughs took place, radically changing the face of computing.

Various areas, such as:

• medical analysis
• autonomous driving
• weather forecast
• the generation of deepfakes

have seen their processes transformed by artificial intelligence, paving the way for new innovations.

A deserved Nobel Prize

Geoffrey Hinton was previously awarded the Turing Prize, often considered the Nobel Prize for computing. Experts agree, however, that this Nobel Prize in Physics celebrates fundamental contributions made to the scientific community, particularly by physicists. The idea that these algorithms have their origins in physical concepts reinforces the importance of their work.

As van der Wilk said, the Nobel Prize represents recognition for the methodological development of artificial intelligence and the essential role of physicists in this field.

The nature of AI

It is crucial to keep in mind that despite the appearance of creativity in systems like ChatGPT, the heart of artificial intelligence relies on mathematical calculations. Hinton himself reminds us that there is “no magic here”: everything comes down to multiplication and addition operations.

Relive the announcement in video: