Page 316 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

References 309

are superior to those that can be developed by hand today; it is also likely to make it
possible to apply deep learning to a wider array of tasks and applications in the
future.

7. CONCLUSION
Evolutionary optimization makes it possible to construct more complex deep
learning architectures than can be done by hand. The topology, components, and
hyperparameters of the architecture can all be optimized simultaneously to fit the
requirements of the task, resulting in superior performance. This automated design
can make new applications of deep learning possible in vision, speech, language, and
other areas. Currently such designs are comparable with best human designs; with
anticipated increases in computing power, they should soon surpass them, putting
the power to good use.

REFERENCES
[1] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to docu-
ment recognition, Proceedings of the IEEE 86 (1998) 22782324.
[2] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Computation 9 (1997)
1735e1780.
[3] R. Collobert, J. Weston, A unified architecture for natural language processing: deep
neural networks with multitask learning, in: Proceedings of the 25th International
Conference on Machine Learning, ACM, 2008, pp. 160e167.
[4] A. Graves, A. Mohamed, G. Hinton, Speech recognition with deep recurrent neural
networks, in: 2013 IEEE International Conference on Acoustics, Speech and Signal
Processing, IEEE, 2013, pp. 6645e6649.
[5] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, Z. Wojna, Rethinking the inception
architecture for computer vision, in: Proceedings of Computer Vision and Pattern
Recognition Conference, 2016, pp. 2818e2826.
[6] Z. Che, S. Purushotham, K. Cho, D. Sontag, Y. Liu, Recurrent Neural Networks for
Multivariate Time Series with Missing Values, 2016. ArXiv 1606.01865.
[7] K. He, X. Zhang, S. Ren, J. Sun, Identity Mappings in Deep Residual Networks, 2016.
ArXiv 1603.05027.
[8] J.Y.-H. Ng, M.J. Hausknecht, S. Vijayanarasimhan, O. Vinyals, R. Monga, G. Toderici,
Beyond Short Snippets: Deep Networks for Video Classification, 2015. ArXiv
1503.08909.
[9] H. Hoos, Programming by optimization, Communications of the ACM 55 (2012) 70e80.
[10] C. Le Goues, T. Nguyen, S. Forrest, W. Weimer, Genprog: automatic bug correction in
real programs, ACM Transactions on Software Engineering 38 (2012).
[11] R. Miikkulainen, N. Iscoe, A. Shagrin, R. Rapp, S. Nazari, P. McGrath, C. Schoolland,
E. Achkar, M. Brundage, J. Miller, J. Epstein, G. Lamba, Sentient Ascend: AI-based
massively multivariate conversion rate optimization, in: Proceedings of Innovative
Applications of Artificial Intelligence, 2018.