no code implementations • 13 Feb 2018 • Masaaki Imaizumi, Kenji Fukumizu
We theoretically discuss why deep neural networks (DNNs) performs better than other models in some cases by investigating statistical properties of DNNs for non-smooth functions.
no code implementations • 1 Aug 2017 • Masaaki Imaizumi, Takanori Maehara, Kohei Hayashi
Tensor train (TT) decomposition provides a space-efficient representation for higher-order tensors.
no code implementations • 31 Jul 2017 • Satoshi Hara, Takayuki Katsuki, Hiroki Yanagisawa, Masaaki Imaizumi, Takafumi Ono, Ryo Okamoto, Shigeki Takeuchi
We show that the proposed method is computationally efficient and does not require any extra computation for model selection.
no code implementations • 19 Jun 2015 • Masaaki Imaizumi, Kohei Hayashi
Nonparametric extension of tensor regression is proposed.
no code implementations • NeurIPS 2017 • Masaaki Imaizumi, Takanori Maehara, Kohei Hayashi
Tensor train (TT) decomposition provides a space-efficient representation for higher-order tensors.
no code implementations • ICML 2017 • Masaaki Imaizumi, Kohei Hayashi
Real data tensors are usually high dimensional but their intrinsic information is preserved in low-dimensional space, which motivates to use tensor decompositions such as Tucker decomposition.
no code implementations • 28 Jan 2019 • Kohei Hayashi, Masaaki Imaizumi, Yuichi Yoshida
In this paper, we study random subsampling of Gaussian process regression, one of the simplest approximation baselines, from a theoretical perspective.
no code implementations • 4 Jul 2019 • Ryumei Nakada, Masaaki Imaizumi
In this study, we prove that an intrinsic low dimensionality of covariates is the main factor that determines the performance of deep neural networks (DNNs).
no code implementations • 15 Oct 2019 • Akiyoshi Sannai, Masaaki Imaizumi, Makoto Kawano
To describe the effect of invariance and equivariance on generalization, we develop a notion of a \textit{quotient feature space}, which measures the effect of group actions for the properties.
no code implementations • 1 Jan 2021 • Manohar Kaul, Masaaki Imaizumi
In this paper, we overcome these obstacles by capturing higher-order interactions succinctly as simplices, model their neighborhood by face-vectors, and develop a nonparametric kernel estimator for simplices that views the evolving graph from the perspective of a time process (i. e., a sequence of graph snapshots).
no code implementations • 4 Nov 2020 • Masaaki Imaizumi, Kenji Fukumizu
We develop a minimax rate analysis to describe the reason that deep neural networks (DNNs) perform better than other standard methods.
no code implementations • 31 Dec 2020 • Masaaki Imaizumi, Taisuke Otsu
This study develops a non-asymptotic Gaussian approximation theory for distributions of M-estimators, which are defined as maximizers of empirical criterion functions.
Statistics Theory Statistics Theory
no code implementations • 5 Feb 2021 • Masatoshi Uehara, Masaaki Imaizumi, Nan Jiang, Nathan Kallus, Wen Sun, Tengyang Xie
We offer a theoretical characterization of off-policy evaluation (OPE) in reinforcement learning using function approximation for marginal importance weights and $q$-functions when these are estimated using recent minimax methods.
no code implementations • 6 Feb 2021 • Manohar Kaul, Masaaki Imaizumi
In this paper, we overcome these obstacles by capturing higher-order interactions succinctly as \textit{simplices}, model their neighborhood by face-vectors, and develop a nonparametric kernel estimator for simplices that views the evolving graph from the perspective of a time process (i. e., a sequence of graph snapshots).
no code implementations • 28 Feb 2021 • Ryumei Nakada, Masaaki Imaizumi
We investigate the asymptotic risk of a general class of overparameterized likelihood models, including deep models.
1 code implementation • 7 Jun 2021 • Rui Zhang, Krikamol Muandet, Bernhard Schölkopf, Masaaki Imaizumi
Kernel maximum moment restriction (KMMR) recently emerges as a popular framework for instrumental variable (IV) based conditional moment restriction (CMR) models with important applications in conditional moment (CM) testing and parameter estimation for IV regression and proximal causal learning.
no code implementations • 3 Aug 2021 • Masahiro Kato, Masaaki Imaizumi, Kenichiro McAlinn, Haruo Kakehi, Shota Yasui
To address this issue, we propose a method that transforms conditional moment restrictions to unconditional moment restrictions through importance weighting, using a conditional density ratio estimator.
no code implementations • ICLR 2022 • Masahiro Kato, Masaaki Imaizumi, Kenichiro McAlinn, Shota Yasui, Haruo Kakehi
We consider learning causal relationships under conditional moment restrictions.
1 code implementation • 7 Nov 2021 • Hikaru Ibayashi, Masaaki Imaizumi
An "escape efficiency" has been an attractive notion to tackle this question, which measures how SGD efficiently escapes from sharp minima with potentially low generalization performance.
no code implementations • 27 Sep 2018 • Masaaki Imaizumi, Kenji Fukumizu
This paper focuses the situation where the target probability measure satisfies the disconnected support property, which means a separate support of a probability, and relates it with the advantage of GANs.
no code implementations • 25 Sep 2019 • Akiyoshi Sannai, Masaaki Imaizumi
Learning problems with data that are invariant to permutations are frequently observed in various applications, for example, point cloud data and graph neural networks.
1 code implementation • 1 Dec 2021 • Akifumi Okuno, Masaaki Imaizumi
The derived minimax rate corresponds to that of the non-invertible bi-Lipschitz function, which shows that the invertibility does not reduce the complexity of the estimation problem in terms of the rate.
1 code implementation • 12 Jan 2022 • Masaaki Imaizumi, Johannes Schmidt-Hieber
We argue that under reasonable assumptions, the local geometry forces SGD to stay close to a low dimensional subspace and that this induces another form of implicit regularization and results in tighter bounds on the generalization error for deep neural networks.
no code implementations • 12 Jan 2022 • Masahiro Kato, Kaito Ariu, Masaaki Imaizumi, Masahiro Nomura, Chao Qin
We show that a strategy following the Neyman allocation rule (Neyman, 1934) is asymptotically optimal when the gap between the expected rewards is small.
no code implementations • 31 Jan 2022 • Masahiro Kato, Masaaki Imaizumi, Kentaro Minami
This paper provides a unified perspective for the Kullback-Leibler (KL)-divergence and the integral probability metrics (IPMs) from the perspective of maximum likelihood density-ratio estimation (DRE).
no code implementations • 10 Feb 2022 • Masahiro Kato, Masaaki Imaizumi
We study the benign overfitting theory in the prediction of the conditional average treatment effect (CATE), with linear regression models.
no code implementations • 18 Apr 2022 • Shogo Nakakita, Masaaki Imaizumi
Then, we derive bounds of risks by the estimator for the cases where the temporal correlation of each coordinate of dependent data is homogeneous and heterogeneous, respectively.
no code implementations • 15 Sep 2022 • Masahiro Kato, Masaaki Imaizumi, Takuya Ishihara, Toru Kitagawa
We then develop the ``Random Sampling (RS)-Augmented Inverse Probability weighting (AIPW) strategy,'' which is asymptotically optimal in the sense that the probability of misidentification under the strategy matches the lower bound when the budget goes to infinity in the small-gap regime.
no code implementations • 6 Feb 2023 • Masahiro Kato, Masaaki Imaizumi, Takuya Ishihara, Toru Kitagawa
We evaluate the decision based on the expected simple regret, which is the difference between the expected outcomes of the best arm and the recommended arm.
no code implementations • 25 May 2023 • Tomoya Wakayama, Masaaki Imaizumi
In the field of high-dimensional Bayesian statistics, a plethora of methodologies have been developed, including various prior distributions that result in parameter sparsity.
no code implementations • 8 Jul 2023 • Masaaki Imaizumi
We show the sup-norm convergence of deep neural network estimators with a novel adversarial training scheme.
no code implementations • 20 Jul 2023 • Masahiro Kato, Akari Ohda, Masaaki Imaizumi, Kenichiro McAlinn
In this paper, we first point out that existing SCMs suffer from an implicit endogeneity problem, which is the correlation between the outcomes of untreated units and the error term in the model of a counterfactual outcome.
no code implementations • 25 Oct 2023 • Masahiro Kato, Masaaki Imaizumi
This study assumes two linear regression models between a potential outcome and covariates of the two treatments and defines CATEs as a difference between the linear regression models.
no code implementations • 30 Jan 2024 • Shuhei Kashiwamura, Ayaka Sakata, Masaaki Imaizumi
However, the selection of quantization hyperparameters, like the number of bits and value range for weight quantization, remains an underexplored area.
1 code implementation • 23 Jun 2021 • Hikaru Ibayashi, Takuo Hamaguchi, Masaaki Imaizumi
Toward achieving robust and defensive neural networks, the robustness against the weight parameters perturbations, i. e., sharpness, attracts attention in recent years (Sun et al., 2020).
2 code implementations • 15 Oct 2020 • Rui Zhang, Masaaki Imaizumi, Bernhard Schölkopf, Krikamol Muandet
We investigate a simple objective for nonlinear instrumental variable (IV) regression based on a kernelized conditional moment restriction (CMR) known as a maximum moment restriction (MMR).
1 code implementation • 30 Jan 2023 • Yuhta Takida, Masaaki Imaizumi, Takashi Shibuya, Chieh-Hsin Lai, Toshimitsu Uesaka, Naoki Murata, Yuki Mitsufuji
Generative adversarial networks (GANs) learn a target probability distribution by optimizing a generator and a discriminator with minimax objectives.
Ranked #1 on Image Generation on FFHQ 1024 x 1024