Sho Sonoda
Sho Sonoda
RIKEN Center for Advanced Intelligence Project (AIP)
Verified email at - Homepage
Cited by
Cited by
Neural network with unbounded activation functions is universal approximator
S Sonoda, N Murata
Applied and Computational Harmonic Analysis 43 (2), 233-268, 2015
Transport Analysis of Infinitely Deep Neural Network
S Sonoda, N Murata
Journal of Machine Learning Research, 2016
Double Continuum Limit of Deep Neural Networks
S Sonoda, N Murata
ICML 2017 Workshop on Principled Approaches to Deep Learning, 1-5, 2017
A statistical model for predicting the liquid steel temperature in ladle and tundish by bootstrap filter
S Sonoda, N Murata, H Hino, H Kitada, M Kano
ISIJ international 52 (6), 1086-1091, 2012
Differentiable multiple shooting layers
S Massaroli, M Poli, S Sonoda, T Suzuki, J Park, A Yamashita, H Asama
Advances in Neural Information Processing Systems 34, 16532-16544, 2021
Neural network with unbounded activations is universal approximator
S Sonoda, N Murata
CoRR, abs/1505.03654, 2015
Learning with optimized random features: Exponential speedup by quantum machine learning without sparsity and low-rank assumptions
H Yamasaki, S Subramanian, S Sonoda, M Koashi
Advances in neural information processing systems 33, 13674-13687, 2020
Sampling hidden parameters from oracle distribution
S Sonoda, N Murata
Artificial Neural Networks and Machine Learning–ICANN 2014: 24th …, 2014
Fully-connected network on noncompact symmetric space and ridgelet transform based on helgason-fourier analysis
S Sonoda, I Ishikawa, M Ikeda
International Conference on Machine Learning, 20405-20422, 2022
Ridge regression with over-parametrized two-layer networks converge to ridgelet spectrum
S Sonoda, I Ishikawa, M Ikeda
International Conference on Artificial Intelligence and Statistics, 2674-2682, 2021
Universality of group convolutional neural networks based on ridgelet analysis on groups
S Sonoda, I Ishikawa, M Ikeda
Advances in Neural Information Processing Systems 35, 38680-38694, 2022
Exponential error convergence in data classification with optimized random features: Acceleration by quantum machine learning
H Yamasaki, S Sonoda
arXiv preprint arXiv:2106.09028, 2021
How powerful are shallow neural networks with bandlimited random weights?
M Li, S Sonoda, F Cao, YG Wang, J Liang
International Conference on Machine Learning, 19960-19981, 2023
The global optimum of shallow neural network is attained by ridgelet transform
S Sonoda, I Ishikawa, M Ikeda, K Hagihara, Y Sawano, T Matsubara, ...
arXiv preprint arXiv:1805.07517v3, 2019
Transportation analysis of denoising autoencoders: a novel method for analyzing deep neural networks
S Sonoda, N Murata
NIPS Workshop on Optimal Transport & Machine Learning, 2017
Lpml: Llm-prompting markup language for mathematical reasoning
R Yamauchi, S Sonoda, A Sannai, W Kumagai
arXiv preprint arXiv:2309.13078, 2023
Ghosts in neural networks: Existence, structure and role of infinite-dimensional null space
S Sonoda, I Ishikawa, M Ikeda
arXiv preprint arXiv:2106.04770, 2021
Quantum ridgelet transform: winning lottery ticket of neural networks with quantum computation
H Yamasaki, S Subramanian, S Hayakawa, S Sonoda
International Conference on Machine Learning, 39008-39034, 2023
EEG dipole source localization with information criteria for multiple particle filters
S Sonoda, K Nakamura, Y Kaneda, H Hino, S Akaho, N Murata, ...
Neural networks 108, 68-82, 2018
Nonparametric weight initialization of neural networks via integral representation
S Sonoda, N Murata
arXiv preprint arXiv:1312.6461, 2013
The system can't perform the operation now. Try again later.
Articles 1–20