Home Page >  News List >> Tech >> Tech

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Tech 2024-10-08 22:40:21 Source: Network
AD

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep LearningThe 2024 Nobel Prize in Physics was awarded to John J. Hopfield and Turing Award winner, AI Godfather Geoffrey E

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

The 2024 Nobel Prize in Physics was awarded to John J. Hopfield and Turing Award winner, AI Godfather Geoffrey E. Hinton, in recognition of their groundbreaking discoveries and inventions in machine learning using artificial neural networks. Hinton expressed surprise in a phone interview, stating that he had not expected it at all.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

John J. Hopfield was born in Chicago, Illinois, in 1933. He received his Bachelor's degree from Swarthmore College in 1954 and his Ph.D. in Physics from Cornell University in 1958. He worked at Bell Labs' Theoretical Group for two years and then taught at the University of California, Berkeley, Princeton University, California Institute of Technology, and Princeton University. He is currently an Emeritus Professor of Molecular Biology at Howard Hughes Medical Institute. He maintained a close connection with Bell Labs for 35 years and was one of the founders of the Ph.D. program in Computational and Neural Systems at Caltech in 1986.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Hopfield's academic achievements primarily focus on condensed matter physics and neuroscience. His most influential papers include: "The Contribution of Excitons to the Complex Dielectric Constant of Crystals" (1958) describing polarons; "Electron transfer between biological molecules by thermally activated tunneling" (1974) describing the quantum mechanics of long-range electron transfer; "Kinetic Proofreading: A New Mechanism for Reducing Errors in Biosynthetic Processes" (1974); "Neural Networks and Physical Systems with Emergent Collective Computational Abilities" (1982) (known as the Hopfield network); and "Neural Computation of Decisions in Optimization Problems" (1985) co-authored with D. W. Tank. His current research and recent publications focus on understanding how the timing and synchrony of neuronal action potentials play a role in biological computation.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Geoffrey E. Hinton, a British-born Canadian computer scientist and psychologist, is a professor at the University of Toronto, known for his contributions to artificial neural networks. He is one of the inventors of the backpropagation algorithm and the Contrastive Divergence algorithm and an active advocate for deep learning, earning the title of "Godfather of Deep Learning." Hinton, along with Yoshua Bengio and Yann LeCun, received the 2018 Turing Award for their contributions to deep learning.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Hinton is a legendary figure, and his family history is equally fascinating. His "great-great-grandfather," George Boole (1815-1864), was a renowned mathematician, after whom Boolean algebra and Boolean logic are named. Boole's wife, Mary, was a feminist philosopher and a self-taught math enthusiast, particularly fond of algebra. Mary's maiden name (Everest) also produced several notable scientists. In fact, Everest is the Western name for Mount Everest, named after Mary's uncle, George Everest, who discovered the mountain while surveying India. Geoffrey Hinton's middle name is Everest, taken from the mountain.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Boole and Mary had five daughters, all of whom, along with their spouses and descendants, achieved remarkable feats. The eldest and youngest of Boole's daughters had connections to China. Boole's eldest daughter, Mary, married Charles Hinton, a mathematician and writer, and had four sons. Among their grandchildren, Han Ding was a "good friend of the Chinese people" and wrote a novel about China's land reform, entitled "Turnover." Han Ding's sister, Han Chun, was a female nuclear physicist who participated in the Manhattan Project, while her husband, Yang Zao, was an American cattle expert. The couple lived in China for many years. Boole's fifth daughter, Ethel Lilian Voynich, is familiar to the Chinese as the author of "The Gadfly."

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Hinton's father was a professor at Cambridge University, a renowned entomologist, and an authority on beetles. Hinton and his three siblings grew up in a house filled with animals, and his father even kept poisonous snakes in a pit in the garage. Hinton's father was a strict man with a high opinion of himself, seemingly believing his son could never reach his heights.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

Hinton's academic journey was not always smooth. He attended what he called a "second-rate public school," Clifton College, for high school. However, Hinton encountered a very intelligent classmate there who told him, "You know, the brain's memory isn't stored in one specific place, but distributed throughout the whole brain, propagating across the entire neural network." His friend then explained, "The brain uses holograms, where you can cut off half, but you still get the whole picture, so..." These words excited Hinton, and it was from this point onward that he became deeply fascinated by the workings of the brain.

AI Godfather Hinton and Neural Networks: From Skepticism to Warning, the Pioneer of Deep Learning

At the age of 18, Hinton entered Kings College, Cambridge, to study Physics, Chemistry, and Mathematics, but dropped out after a month. He tried various disciplines, including architecture, philosophy, physics, and physiology, but ultimately failed to persevere. Finally, in 1970, he received a Bachelor's degree in experimental psychology and then abandoned Cambridge University to become a carpenter. While making bookshelves and wooden doors, he pondered the workings of the human brain, believing it to be his preferred way of life.

After more than a year, Hinton had a new idea. He decided to return to academia and try a new direction: artificial intelligence. In 1972, he enrolled for a Ph.D. at Edinburgh University in Scotland, where he met Professor Higgins (Christopher Higgins), who was researching neural networks. However, just as he began pursuing this goal, Professor Higgins changed his academic focus, shifting towards the symbolic side of AI, believing connectionist neural networks to be nonsense.

In the late 1970s, there was a group in California, San Diego, particularly David Rumelhart, who found neural networks incredibly interesting. In 1986, Hinton, together with David Rumelhart and Ronald Williams, published a paper entitled "Learning representations by back-propagating errors." They applied the backpropagation algorithm to multi-layer neural networks and proved its effectiveness for machine learning. Their paper also demonstrated that multiple hidden layers in neural networks could learn any function, thus solving the problem of single-layer perceptrons raised in books by Minsky and others. During the same period, Hinton, along with David Ackley and Terry Sejnowski, invented the Boltzmann machine.

Hinton found a job at Carnegie Mellon University in Pittsburgh, but he grew increasingly uneasy about American foreign policy under Reagan, especially its interference in Central America. He and his wife, Rose, adopted a boy and a girl from South America but didn't feel comfortable raising them in the United States. Additionally, most AI research in the United States was funded by the Department of Defense, which also dissatisfied Hinton. Therefore, he accepted an invitation from the Canadian Institute for Advanced Research and a position in the Computer Science department at the University of Toronto, starting the "Machine and Brain Learning" project in Canada.

Hinton's two important articles in 1986 on backpropagation algorithms and Boltzmann machines failed to withstand the "AI winter" and seemed to have little impact. However, Hinton secured a stable position at the University of Toronto in Canada and received sufficient research funding for neural networks, enabling him to relentlessly pursue this neglected field for over 30 years without regret. As time passed, some believers in deep learning were drawn to Hinton. He cultivated many students, who in turn had their own students, along with postdocs and collaborators. The talent pool researching neural network deep learning was filled with shining stars. Despite the limited job opportunities and scarce funding during the winter, these researchers were enthusiastic. They relied on their own beliefs, ignored the noisy distractions, and found joy in their work. The world appeared calm on the surface, but an undercurrent surged beneath, preparing for the arrival of spring for artificial intelligence.

Hinton's long-term cultivation in this neglected field ultimately ushered in spring, not only earning him the 2018 Turing Award but also bringing revolutionary breakthroughs to the field of artificial intelligence. However, the AI guru hesitated, harboring doubt, even anger, towards the machine he taught to learn. He feared that his lifelong work might lead to the end of humanity, believing his ultimate mission was to warn the world!

Hinton's experiences and his warnings about artificial intelligence remind us that technological development needs to be synchronized with ethics and social responsibility. While enjoying the convenience and progress brought by artificial intelligence, we must also contemplate how to avoid its negative impacts and ensure that it truly serves the well-being of humanity.


Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])

Mobile advertising space rental

Tag: AI Godfather Hinton and Neural Networks From Skepticism to

Guess you like

Unite directoryCopyright @ 2011-2024 All Rights Reserved. Copyright Webmaster Search Directory System