1. [1] S. Amari, “Any target function exists in a neighborhood of any sufficiently wide random network: A geometrical perspective,” Neural Computation, vol. 32, no. 8, pp. 1431-1447, 2020. DOI: 10.1162/neco_a_01295
2. [2] S. Amari, “mathematical engineering and IT,” IEICE ICT Pioneers Webinar Series, (in Japanese on-demand video on the trial archive), September 24, 2020. https://webinar.ieice.org/summary.php?id=175&expandable=0&code=PNS&sel=&year=2020
3. [3] A. Jacot, F. Gabriel, and C. Hongler, “Neural tangent kernel: Convergence and generalization in neural networks,” Advances in Neural Information Processing Systems 32, S. Bengio, and H.M. Wallach (eds), pp. 8571-8580, 2018.
4. [4] J. Lee, L. Xiao, S.S. Schoenholz, Y. Bahri, R. Novak, J. Sohl-Dickstein, and J. Pennington, “Wide neural networks of any depth evolve as linear models under gradient descent,” Advances in Neural Information Processing Systems 31, H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché Buc, E. Fox, and R. Garnett (eds), pp. 8572-8583, 2019.
5. [5] M. Mézard and A. Montanari, Information, Physics, and Computation, OXFORD University Press, New York, 2009.