# Geoffrey Hinton

British-Canadian computer scientist and cognitive psychologist, known as a pioneer of deep learning and backpropagation, awarded the 2024 Nobel Prize in Physics for foundational work on neural networks.

Geoffrey Everest Hinton (born 6 December 1947) is a British-Canadian computer scientist, cognitive scientist, and cognitive psychologist known for his foundational work on artificial neural networks, which earned him the title 'the Godfather of AI'. He is University Professor Emeritus at the [University of Toronto](https://www.wikiprompt.org/wiki/university-of-toronto), where he has been a central figure in the development of deep learning since the late 1980s.

Hinton's contributions include co-authorship of the landmark 1986 paper that popularized backpropagation for training multi-layer neural networks, and the design of AlexNet, which won the 2012 ImageNet challenge and revolutionized computer vision. He shared the 2018 Turing Award with Yoshua Bengio and Yann LeCun for their work on deep learning, and in 2024 he received the Nobel Prize in Physics with John Hopfield for their discoveries enabling machine learning with artificial neural networks.

## Education

Hinton was born in Wimbledon, England, and attended Clifton College in Bristol, before matriculating at King's College, Cambridge in 1967. After trial across fields including natural sciences and history of art, he graduated with a Bachelor of Arts in experimental psychology in 1970. He then worked as an apprentice carpenter, but returned to academia, studying from 1972 to 1975 at the University of Edinburgh, where he received a PhD in artificial intelligence in 1978, supervised by Christopher Longuet-Higgins. His early interest in neural networks diverged from the then-dominant symbolic AI school.

## Career and academic roles

After completing his PhD, Hinton held positions at the University of Sussex and the MRC Applied Psychology Unit in the UK, but faced difficulty securing research funding. He moved to the United States, working at the University of California, San Diego, and later at Carnegie Mellon University in Pittsburgh. In 1987, he joined the University of Toronto, where he has remained ever since, with a break from 1998 to 2001 when he served as founding director of the Gatsby Charitable Foundation Computational Neuroscience Unit at University College London.

At the [University of Toronto](https://www.wikiprompt.org/wiki/university-of-toronto), Hinton was appointed a fellow of the Canadian Institute for Advanced Research (CIFAR) in 1987, and in 2004 he helped launched a CIFAR program on Neural Computation and Adaptive Perception, later renamed Learning in Machines & Brains, which he led for ten years. That program's membership included Yoshua Bengio and Yann LeCun, with whom he later shared the Turing Award. In 2012, Hinton taught a free online course on [neural networks](https://www.wikiprompt.org/wiki/neural-network) through the Coursera platform, introducing hundreds of thousands of people to [deep learning](https://www.wikiprompt.org/wiki/deep-learning). That same year, he co-founded DNNresearch Inc. with his graduate students Alex Krizhevsky and Ilya Sutskever.

The company was acquired by [Google](https://www.wikiprompt.org/wiki/google-deepmind) (later [Google Brain](https://www.wikiprompt.org/wiki/google-brain)) in March 2013 for $44 million, and Hinton then split his time between the company and his academic duties. In May 2023, he publicly resigned from Google, saying he wanted to speak freely about the risks of AI, and that he partly regretted his life's work. He also co-founded and became chief scientific advisor of the Vector Institute in Toronto in 2017, a position he continued after his retirement from active research. Hinton is also University Professor Emeritus in the Department of Computer Science, with doctoral students and postdoctoral researchers including Peter Dayan, Sam Roweis, Max Welling, Richard Zemel, Brendan Frey, Radford Neal, Yee Whye Teh, Ruslan Salakhutdinov, Ilya Sutskever, Yann LeCun, and Zoubin Ghahramani.

## Research in neural networks

Hinton's research primarily concerns the use of [neural networks](https://www.wikiprompt.org/wiki/neural-network) for machine learning, memory, perception, and symbol processing. He has authored or co-authored more than 200 peer-reviewed publications, and his work spans distributed representations, time-delay neural networks, mixtures of experts, Helmholtz machines, and products of experts.

In the 1980s, Hinton was a member of the Parallel Distributed Processing group at [Carnegie Mellon University](https://www.wikiprompt.org/wiki/carnegie-mellon-university), which also included Terrence Sejnowski, Francis Crick, David Rumelhart, and James McClelland. This group championed connectionism, the idea that higher-level cognition could emerge from networks of simple units, during a period when the AI field was largely orienting toward symbolic AI, an approach explicitly programming rules and knowledge.

A key technological contribution came in 1985 when Hinton co-invented the Boltzmann machine with David Ackley and Terry Sejnowski. This stochastic network learned to represent and generate patterns via a laten probability model, laying groundwork for later energy-based models. In 1995, with his students and colleagues, Hinton proposed the wake-sleep algorithm, which trains a recognition network and a generative network using alternating 'wake' and 'sleep' phases. That work contributed to later advances in several learning approaches.

Hinton also developed visualization tricks that used to understand high-dimensional data. In 2008, with Laurens van der Maaten, he introduced t-SNE, a method that reduces data from many dimensions to two or three for visual inspection, widely used in fields beyond neural networks. He also coauthored a 2007 paper on unsupervised learning of image transformations, furthericheting the toolkit for learning from data without labels.

## AlexNet and the deep learning breakthrough

Hinton's prominence surged in 2012 when AlexNet, a convolutional neural network designed with his students Alex Krizhevsky and Ilya Sutskever, decisively won the ImageNet Large Scale Visual Recognition Challenge. The network achieved top-5 error rate of 15.3 percent at the time, far better than previous methods, dramatically showing the power of deep neural networks. This breakthrough sparked a renaissance in [deep learning](https://www.wikiprompt.org/wiki/deep-learning) computer [computer vision](https://www.wikiprompt.org/wiki/computer-vision) and was a key factor in the current wave of [AI](https://www.wikiprompt.org/wiki/artificial-intelligence) advances.

Before AlexNet, neural networks had been considered limited, often trained on small datasets. The success combined large [data augmentation](https://www.wikiprompt.org/wiki/data-augmentation), rectified linear units, and efficient GPU training, all of which Hinton and his colleagues had pursued for years. The recognition of AlexNet's performance is widely seen as a major motivator for why new companies and research labs like Google's AI efforts stepped up their investment [deep-learning](https://www.wikiprompt.org/wiki/deep-learning).

## Public outreach and collaborations

Hinton has consistently worked to communicate his ideas to a broader public, publishing writing on [artificial intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence) in magazines like Scientific American in September 1992 and October 1993. He also taught the 2012 MOOC on [learning the real knowledge](https://www.wikiprompt.org/wiki/neural-network) which helped spawn a generation of deep learning researchers. His move to Google in 2013, announced as a part-time appointment meant to \"divide my time between university research and work at Google,\" further instantiated tie to industry. After founding DNNresearch with Suverskever and Krizhevsky, he joined the ecosystem of AI companies, most notable later that his group collaborated with [openai](https://www.wikiprompt.org/wiki/openai) early efforts like the 2020 language model differences, but Hinton himself was not at OpenAI. He has staff members now at various institutions.

## Award and recognition

Hinton's foundational discoveries earned him the highest professional recognitions for his field. In 2018, he received the ACM AM Turing Award, often called the Nobel Prize of computing, jointly with Yoshua Bengio and Yann LeCun, in recognition of conceptual and engineering breakthrough that proved a notion that large neural networks could yield intelligence. They are often called the 'Godfathers of Deep Learning'.

In 2024, Hinton was awarded the Nobel Prize in Physics along with John Hopfield for 'for foundational discoveries and inventions that enable machine learning with artificial neural-networks'. The Swedish Academy highlighted their work on statistical physics and associative memory, which laid the major foundations for the current boom in [artificial-intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence). Hinton, a theoretical physicist by joint training, accepted the prize, reflecting on the irony that a field once seen as marginal became central to modern physics.

## AI safety and advocacy

After leaving Google in 2023, Hinton became a prominent voice concerning the risks of AI. He has stated that he left to value is to speak about artificial-safety. He says he worries about [large-language-model](https://www.wikiprompt.org/wiki/large-language-model)s and other systems that may soon function at a level human capacity, and they could at best be misused by malicious actors or at worst become existential risks. He has no longer considered that transformation from [artificial-general-intelligence](https://www.wikiprompt.org/wiki/artificial-general-intelligence) is possible, but huge accelerating.

Hinton has called for establishing and enforcing safety guidelines through binding cooperation among AI companies and nations. He has stated that AI self-regulation is insufficient, and has specifically highlighted that companies building [large-language-model](https://www.wikiprompt.org/wiki/large-language-model)s may be "enslaving" through profit incentives, and that nations could compete in a race that could be avoidable to safety. After receiving the Nobel Prize, he proposed urgent research on AI safety to design systems to ensure control of those smarter than humans. His message received for scientific criticism but resonates with the idea of an AI community increasingly concerned about control.

## Personal life and legacy

Hinton was born in Wimbledon, London, into an intellectual family. A background includes many scientists, and his great-grandfather Claude was a logician. He emigrated to Canada in 1987, became a Canadian citizen, and he now lives in Toronto. He used eccentricities like his famous custom hacking and has used to have no shortage of success. His early visionary contributions, often mocked in the 1980s, were subsequently shown as prescient: It\n
Notable former PhD students and postdoctoral researchers of Hinton include Peter Stimson, Sam Read, Max Welles, Richard Urban Bedding, Radford, Y. Why, Ruslan, Ilya Sutskever (co-founder of OpenAI, creator of the [deep-learning](https://www.wikiprompt.org/wiki/deep-learning) at)\n
Hinton has been recognized as a foundational figure insuring H that his life's work. He has called for ethical measures, but also continue to see value in the AI \"era of intellectual transformation.\"

---
Source: https://www.wikiprompt.org/wiki/geoffrey-hinton-3
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-10-07T16:49:25.468019+00:00