Geoffrey E. Hinton

OpenAlex profile · ORCID unavailable

Selected research works

Up to 100 selected research works. Source updated 2026-10-28T13:16:11+00:00.

1. ImageNet classification with deep convolutional neural networks · 2017 · 112,170 citations
2. Deep learning · 2015 · 84,078 citations
3. Dropout: a simple way to prevent neural networks from overfitting · 2014 · 33,420 citations
4. Visualizing Data using t-SNE · 2008 · 33,051 citations
5. Learning representations by back-propagating errors · 1986 · 31,873 citations
6. Reducing the Dimensionality of Data with Neural Networks · 2006 · 20,929 citations
7. Learning Internal Representations by Error Propagation · 1985 · 16,525 citations
8. A Fast Learning Algorithm for Deep Belief Nets · 2006 · 16,335 citations
9. Distilling the Knowledge in a Neural Network · 2015 · 14,202 citations
10. Rectified Linear Units Improve Restricted Boltzmann Machines · 2010 · 13,997 citations
11. A Simple Framework for Contrastive Learning of Visual Representations · 2020 · 10,665 citations
12. Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups · 2012 · 10,173 citations
13. Speech recognition with deep recurrent neural networks · 2013 · 8,892 citations
14. Improving neural networks by preventing co-adaptation of feature detectors · 2012 · 6,382 citations
15. Adaptive Mixtures of Local Experts · 1991 · 5,321 citations
16. Training Products of Experts by Minimizing Contrastive Divergence · 2002 · 4,880 citations
17. Dynamic Routing Between Capsules · 2017 · 4,080 citations
18. On the importance of initialization and momentum in deep learning · 2013 · 3,740 citations
19. A Learning Algorithm for Boltzmann Machines* · 1985 · 3,405 citations
20. Layer Normalization · 2016 · 3,266 citations
21. Learning Internal Representations by Error Propagation · 1988 · 3,003 citations
22. A Practical Guide to Training Restricted Boltzmann Machines · 2012 · 2,852 citations
23. Phoneme recognition using time-delay neural networks · 1989 · 2,642 citations
24. A View of the Em Algorithm that Justifies Incremental, Sparse, and other Variants · 1998 · 2,176 citations
25. Deep Boltzmann machines · 2009 · 1,886 citations
26. Restricted Boltzmann machines for collaborative filtering · 2007 · 1,885 citations
27. Neighbourhood Components Analysis · 2004 · 1,857 citations
28. Deep Neural Networks for Acoustic Modeling in Speech Recognition · 2012 · 1,818 citations
29. Acoustic Modeling Using Deep Belief Networks · 2011 · 1,727 citations
30. Stochastic Neighbor Embedding · 2002 · 1,635 citations
31. Connectionist learning procedures · 1989 · 1,494 citations
32. Big Self-Supervised Models are Strong Semi-Supervised Learners · 2020 · 1,431 citations
33. Transforming Auto-Encoders · 2011 · 1,360 citations
34. Improving deep neural networks for LVCSR using rectified linear units and dropout · 2013 · 1,310 citations
35. Generating Text with Recurrent Neural Networks · 2011 · 1,268 citations
36. Semantic hashing · 2008 · 1,247 citations
37. New types of deep neural network learning for speech recognition and related applications: an overview · 2013 · 1,224 citations
38. The Helmholtz Machine · 1995 · 1,196 citations
39. Autoencoders, Minimum Description Length and Helmholtz Free Energy · 1993 · 1,173 citations
40. The "Wake-Sleep" Algorithm for Unsupervised Neural Networks · 1995 · 1,071 citations
41. Learning and Relearning in Boltzmann Machines · 2001 · 1,037 citations
42. Distributed representations · 2018 · 1,025 citations
43. Learning multiple layers of representation · 2007 · 1,013 citations
44. Keeping the neural networks simple by minimizing the description length of the weights · 1993 · 959 citations
45. Parallel models of associative memory · 1981 · 927 citations
46. Matrix capsules with EM routing · 2018 · 910 citations
47. Backpropagation and the brain · 2020 · 907 citations
48. Deep belief networks · 2009 · 907 citations
49. When Does Label Smoothing Help? · 2019 · 859 citations
50. A Scalable Hierarchical Distributed Language Model · 2008 · 827 citations
51. Learning distributed representations of concepts. · 1989 · 824 citations
52. A learning algorithm for boltzmann machines · 1985 · 767 citations
53. Learning Representations by Back-Propagating Errors · 2002 · 760 citations
54. Similarity of Neural Network Representations Revisited · 2019 · 724 citations
55. A general framework for parallel distributed processing · 1986 · 709 citations
56. Modeling Human Motion Using Binary Latent Variables · 2007 · 708 citations
57. How Learning Can Guide Evolution. · 1987 · 681 citations
58. Deep learning for AI · 2021 · 676 citations
59. Schemata and Sequential Thought Processes in PDP Models · 1988 · 665 citations
60. Deep Learning—A Technology With the Potential to Transform Health Care · 2018 · 663 citations
61. Lesioning an attractor network: Investigations of acquired dyslexia. · 1991 · 656 citations
62. Regularizing Neural Networks by Penalizing Confident Output Distributions · 2017 · 655 citations
63. Learning to Detect Roads in High-Resolution Aerial Images · 2010 · 646 citations
64. On Contrastive Divergence Learning. · 2005 · 631 citations
65. Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer · 2017 · 622 citations
66. Classical and Bayesian Inference in Neuroimaging: Theory · 2002 · 603 citations
67. Proceedings of the 1988 Connectionist Models Summer School · 1989 · 602 citations
68. The EM algorithm for mixtures of factor analyzers · 1996 · 596 citations
69. A time-delay neural network architecture for isolated word recognition · 1990 · 596 citations
70. Simplifying Neural Networks by Soft Weight-Sharing · 1992 · 595 citations
71. Machine learning for aerial image labeling · 2013 · 579 citations
72. Three new graphical models for statistical language modelling · 2007 · 569 citations
73. How Neural Networks Learn from Experience · 1992 · 550 citations
74. Feudal Reinforcement Learning · 1992 · 530 citations
75. A Simple Way to Initialize Recurrent Networks of Rectified Linear Units · 2015 · 522 citations
76. CONNECTIONIST LEARNING PROCEDURES11This chapter appeared in Volume 40 of Artificial Intelligence in 1989, reprinted with permission of North-Holland Publishing. It is a revised version of Technical Report CMU-CS-87-115, which has the same title and was prepared in June 1987 while the author was at Carnegie Mellon University. The research was supported by contract N00014-86-K-00167 from the Office of Naval Research and by grant IST-8520359 from the National Science Foundation. · 1990 · 516 citations
77. OPTIMAL PERCEPTUAL INFERENCE · 1983 · 499 citations
78. Self-organizing neural network that discovers surfaces in random-dot stereograms · 1992 · 496 citations
79. Parameter estimation for linear dynamical systems · 1996 · 492 citations
80. An Efficient Learning Procedure for Deep Boltzmann Machines · 2012 · 484 citations
81. Grammar as a Foreign Language · 2014 · 464 citations
82. Application of Deep Belief Networks for Natural Language Understanding · 2014 · 461 citations
83. On rectified linear units for speech processing · 2013 · 442 citations
84. Learning a Nonlinear Embedding by Preserving Class Neighbourhood Structure · 2007 · 438 citations
85. Learning to combine foveal glimpses with a third-order Boltzmann machine · 2010 · 433 citations
86. Replicated Softmax: an Undirected Topic Model · 2009 · 425 citations
87. Exponential Family Harmoniums with an Application to Information Retrieval · 2004 · 424 citations
88. Variational Learning for Switching State-Space Models · 2000 · 423 citations
89. Evaluation of gaussian processes and other methods for non-linear regression · 1997 · 412 citations
90. SMEM Algorithm for Mixture Models · 2000 · 411 citations
91. Distilling a Neural Network Into a Soft Decision Tree · 2017 · 407 citations
92. Using very deep autoencoders for content-based image retrieval. · 2011 · 387 citations
93. Binary coding of speech spectrograms using a deep auto-encoder · 2010 · 378 citations
94. The Appeal of Parallel Distributed Processing · 1988 · 376 citations
95. Glove-Talk: a neural network interface between a data-glove and a speech synthesizer · 1993 · 374 citations
96. Training recurrent neural networks · 2013 · 369 citations
97. The Recurrent Temporal Restricted Boltzmann Machine · 2008 · 367 citations
98. Learning to Label Aerial Images from Noisy Data · 2012 · 363 citations
99. Factored conditional restricted Boltzmann Machines for modeling motion style · 2009 · 362 citations
100. Parallel visual computation · 1983 · 356 citations