Geoffrey Everest Hinton, born on December 6, 1947, is a British born Canadian computer scientist and psychologist, and professor at the University of Toronto. Renowned for its contributions in the field of neural networks. Hinton is one of the inventors of backpropagation algorithm and contrastive divergence algorithm, and an active promoter of deep learning, known as the “father of deep learning”. Hinton was awarded the 2018 Turing Award along with Yoshua Bengio and Yann LeCun for their contributions to deep learning.
Educational Background
Hinton obtained a Bachelor’s degree in Experimental Psychology from the University of Cambridge in the United Kingdom in 1970. Afterwards, he obtained a PhD in Artificial Intelligence from the University of Edinburgh in 1978. Afterwards, he worked at the University of Sussex, University of California, San Diego, University of Cambridge, Carnegie Mellon University, and University College London. He is the founder of the Gatsby Center for Computational Neuroscience and currently serves as a professor in the Department of Computer Science at the University of Toronto. Hinton is the chief scholar in the field of machine learning in Canada and the leader of the “Neural Computing and Adaptive Perception” project sponsored by the Canadian Institute for Advanced Research. Hinton joined Google in March 2013, and Google acquired DNNresearch, the company he founded. You can learn more about it on Google Research.
research results
Hinton’s research investigated methods of using neural networks for machine learning, memory, perception, and symbol processing, and he has authored or co authored over 200 peer-reviewed publications. At the 2022 Neural Information Processing Systems Conference (NeurIPS), Hinton introduced a new neural network learning algorithm, which he called the “Forward Forward” algorithm. The idea of the new algorithm is to replace the traditional forward backward path of backpropagation with two forward paths, one with positive (i.e. real) data and the other with negative data, which can be generated by the network itself.
During Hinton’s tenure as a professor at Carnegie Mellon University (1982-1987), David E. Rumelhart, Hinton, and Ronald J. Williams applied backpropagation algorithms to multi-layer neural networks. Their experiments show that this network can learn useful internal representations of data. In a 2018 interview, Hinton stated that “David E. Rumelhart proposed the fundamental idea of backpropagation, so this is his invention. ”Although this work is important for promoting backpropagation, he is not the first person to propose this method. Seppo Linnainmaa proposed reverse mode automatic differentiation in 1970, with backpropagation being a special case. Paul Werbos proposed using it to train neural networks in 1974.
During the same period, Hinton co invented the Boltzmann machine with David Ackley and Terry Sejnowski. His other contributions to neural network research include distributed representation, delayed neural networks, expert hybrid systems, and Helmholtz machines. In 2007, Hinton co authored an unsupervised learning paper titled ‘Unsupervised Learning for Image Conversion’. Hinton’s research brief can be found in his articles published in Scientific American in September 1992 and October 1993.
Awards and Honors
Fellow of the American Association for Artificial Intelligence (1990)
Fellow of the Royal Society (1998)
Rumelhart Prize (2001)
IJCAI Outstanding Research Award (2005)
IEEE Frank Rosenblatt Award (2014)
James Clark Maxwell Medal (2016)
BBVA Foundation Knowledge Frontier Award (2016)
Turing Award (2018)
Prince of Asturias Award (2022)
相关文章
