no code implementations • 22 Mar 2024 • Yoshihide Sawada, Ryuji Saiin, Kazuma Suetake
Recently, the number of parameters in DNNs has explosively increased, as exemplified by LLMs (Large Language Models), making inference on small-scale computers more difficult.
no code implementations • 14 Mar 2024 • Naoki Hayashi, Yoshihide Sawada
In this paper, we reveal the Bayesian generalization error in PCBM with a three-layered and linear architecture.
no code implementations • 16 Feb 2024 • Fusataka Kuniyoshi, Yoshihide Sawada
This resonance, caused by the interaction between motor and tire vibrations, puts excessive loads on the vehicle's drive shaft.
no code implementations • 2 Dec 2023 • Takashi Furuya, Satoshi Okuda, Kazuma Suetake, Yoshihide Sawada
This instability problem comes from the difficulty of the minimax optimization, and there have been various approaches in GANs and UDAs to overcome this problem.
no code implementations • 4 Oct 2023 • Ryuji Saiin, Tomoya Shirakawa, Sota Yoshihara, Yoshihide Sawada, Hiroyuki Kusumoto
Our proposed method can solve these problems; namely, SAF can halve the number of operations during the forward process, and it can be theoretically proven that SAF is consistent with the Spike Representation and OTTT, respectively.
no code implementations • 16 Mar 2023 • Naoki Hayashi, Yoshihide Sawada
However, it has not yet been possible to understand the behavior of the generalization error in CBM since a neural network is a singular statistical model in general.
no code implementations • 3 Feb 2023 • Kazuma Suetake, Takuya Ushimaru, Ryuji Saiin, Yoshihide Sawada
Spiking neural networks (SNNs) are energy-efficient neural networks because of their spiking nature.
no code implementations • 20 Jun 2022 • Yoshihide Sawada, Keigo Nakamura
In this study, we use a self-explaining neural network (SENN), which learns unsupervised concepts, to acquire concepts that are easy for people to understand automatically.
no code implementations • 3 Mar 2022 • Shin-ichi Ikegawa, Ryuji Saiin, Yoshihide Sawada, Naotake Natori
Biologically inspired spiking neural networks (SNNs) are widely used to realize ultralow-power energy consumption.
1 code implementation • 3 Feb 2022 • Yoshihide Sawada, Keigo Nakamura
We refer to the proposed model as the concept bottleneck model with additional unsupervised concepts (CBM-AUC).
no code implementations • 26 Jan 2022 • Kazuma Suetake, Shin-ichi Ikegawa, Ryuji Saiin, Yoshihide Sawada
To solve these problems, we propose a single-step spiking neural network (S$^3$NN), an energy-efficient neural network with low computational cost and high precision.
1 code implementation • 25 Oct 2019 • Yoshihide Sawada, Koji Morikawa, Mikiya Fujii
Generative models based on generative adversarial networks (GANs) and variational autoencoders (VAEs) have been widely studied in the fields of image generation, speech generation, and drug discovery, but, only a few studies have focused on the generation of inorganic materials.
1 code implementation • 3 Sep 2019 • Jordan Hoffmann, Louis Maestrati, Yoshihide Sawada, Jian Tang, Jean Michel Sellier, Yoshua Bengio
We present a method to encode and decode the position of atoms in 3-D molecules from a dataset of nearly 50, 000 stable crystal unit cells that vary from containing 1 to over 100 atoms.
no code implementations • 19 Apr 2018 • Yoshihide Sawada
We introduce a method to disentangle controllable and uncontrollable factors of variation by interacting with the world.
no code implementations • 13 Nov 2017 • Yoshihide Sawada, Yoshikuni Sato, Toru Nakada, Kei Ujimoto, Nobuhiro Hayashi
One of the conventional methods for solving this problem is transfer learning for DNNs.
no code implementations • CVPR 2013 • Hidekata Hontani, Yuto Tsunekawa, Yoshihide Sawada
In this paper, we propose a new non-rigid robust registration method that registers a point distribution model (PDM) of a surface to given 3D images.