Decoupled neural interfaces using synthetic gradients

34th International Conference on Machine Learning, ICML 2017

Volumn 4, Issue , 2017, Pages 2558-2577

  Jaderberg, Max ^a   Czarnecki, Wojciech Marian ^a   Osindero, Simon ^a   Vinyals, Oriol ^a   Graves, Alex ^a   Silver, David ^a   Kavukcuoglu, Koray ^a  

^a DEEPMIND   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

DIRECTED GRAPHS; ERRORS; FORECASTING; LEARNING SYSTEMS;

ERROR SIGNAL; FEED FORWARD; LOCAL INFORMATION; MODELLING ERROR; NETWORK GRAPH; NEURAL INTERFACES; RECURRENT NEURAL NETWORK (RNNS); WEIGHT UPDATE;

RECURRENT NEURAL NETWORKS;

EID: 85048381780     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (16)

1. Baiduzzi, D, Vanchinathan, H, and Buhmann, J. Kickback cuts backprop's red-tape: Biologically plausible credit assignment in neural networks. arXiv preprint arXiv:1411.6191, 2014.
2. Cooijmans, T., Bailas, N., Laurent, C, and Courville, A. Recurrent batch normalization. arXiv preprint arXiv:1603.09025, 2016.
3. Czarnecki, W M, Swirszcz, G, Jaderberg, M, Osindero, S, Vinyals, O, and Kavukcuoglu, K. Understanding synthetic gradients and decoupled neural interfaces. arXiv preprint, 2017.
4. Fairbank, M. Value-gradient learning. PhD thesis, City University London, UK, 2014.
5. Glorot, X., Bordes, A., and Bengio, Y. Deep sparse rectifier neural networks. In International Conference on Artificial Intelligence and Statistics, pp. 315-323, 2011.
6. Heess, N, Wayne, G, Silver, D, Lillicrap, T P, Erez, T, and Tassa, Y. Learning continuous control policies by stochastic value gradients. In Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015, December 7-12, 2015, Montreal, Quebec, Canada, pp. 2944-2952, 2015.
7. Ioffe, S. and Szegedy, C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. ICML, 2015.
8. Kingma, D. and Ba, J. Adam: A method for stochastic optimization. arXiv preprint arXiv: 1412.6980, 2014.
9. Krizhevsky, A. and Hinton, G. Learning multiple layers of features from tiny images, 2009.
10. LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278-2324, 1998a.
11. LeCun, Y., Cortes, C, and Burges, C. The mnist database of handwritten digits, 1998b.
12. Lillicrap, T P, Cownden, D, Tweed, D B, and Akerman, C J. Random synaptic feedback weights support error back-propagation for deep learning. Nature Communications, 7, 2016.
13. Nokland, A. Direct feedback alignment provides learning in deep neural networks. In Lee, D. D., Sugiyama, M., Luxburg, U. V., Guyon, I., and Garnett, R. (eds.), Advances in Neural Information Processing Systems 29, pp. 1037-1045. Curran Associates, Inc., 2016.
14. Sutton, R S. Learning to predict by the methods of temporal differences. Machine Learning, 3:9-44, 1988.
15. Werbos, P J. Approximating dynamic programming for real-time control and neural modeling. In White, David A. and Sofge, Donald A. (eds.), Handbook of Intelligent Control, chapter 13, pp. 493-525. Van Nostrand Reinhold, New York, 1992.
16. Williams, R. J. and Zipser, D. A learning algorithm for continually running fully recurrent neural networks. Neural computation, 1(2):270-280, 1989.