Zobrazeno 1 - 10
of 994
pro vyhledávání: '"Tsuzuku A"'
Autor:
Wood, David, Lublinsky, Boris, Roytman, Alexy, Singh, Shivdeep, Adam, Constantin, Adebayo, Abdulhamid, An, Sungeun, Chang, Yuan Chi, Dang, Xuan-Hong, Desai, Nirmit, Dolfi, Michele, Emami-Gohari, Hajar, Eres, Revital, Goto, Takuya, Joshi, Dhiraj, Koyfman, Yan, Nassar, Mohammad, Patel, Hima, Selvam, Paramesvaran, Shah, Yousaf, Surendran, Saptha, Tsuzuku, Daiki, Zerfos, Petros, Daijavad, Shahrokh
Data preparation is the first and a very important step towards any Large Language Model (LLM) development. This paper introduces an easy-to-use, extensible, and scale-flexible open-source data preparation toolkit called Data Prep Kit (DPK). DPK is a
Externí odkaz:
http://arxiv.org/abs/2409.18164
Autor:
Kazuaki Oyake, PT, PhD, Shota Watanabe, OT, MSc, Ayano Takeuchi, ST, BSc, Taiki Yoshida, OT, PhD, Takashi Shigematsu, MD, PhD, Yuuki Natsume, ST, BSc, Shigeki Tsuzuku, MD, PhD, Kunitsugu Kondo, MD, PhD, Ichiro Fujishima, MD, PhD, Yohei Otaka, MD, PhD, Satoshi Tanaka, PhD
Publikováno v:
Archives of Rehabilitation Research and Clinical Translation, Vol 6, Iss 3, Pp 100344- (2024)
Objective: To investigate the feasibility of poststroke interventions using a motivational instructional design model with occupational therapy (OT) and swallowing therapy (ST) and the model's potential physical and mental health effects. Design: An
Externí odkaz:
https://doaj.org/article/0870428cd62e4bc6bbb2d9cfa5c852c8
Autor:
Oyake, Kazuaki, Watanabe, Shota, Takeuchi, Ayano, Yoshida, Taiki, Shigematsu, Takashi, Natsume, Yuuki, Tsuzuku, Shigeki, Kondo, Kunitsugu, Fujishima, Ichiro, Otaka, Yohei, Tanaka, Satoshi
Publikováno v:
In Archives of Rehabilitation Research and Clinical Translation September 2024 6(3)
Autor:
Kimura, Akihiro, Tsuzuku, Hanayo
Time-dependent renormalization was employed to derive a nonlinear quantum master equation (QME), in which the dynamics of a non-equilibrium fluctuation in an irrelevant system are fed back into that of a relevant one. In terms of application, the non
Externí odkaz:
http://arxiv.org/abs/2001.00344
Autor:
Atsushi Ishihara, Takashi Yoshizane, Teruki Mori, Yui Sasaki, Takahiro Hosokawa, Jun Suzuki, Akifumi Tsuzuku, Fumihiro Asano, Toshiyuki Noda
Publikováno v:
International Journal of Emergency Medicine, Vol 16, Iss 1, Pp 1-7 (2023)
Abstract Background This study aimed to understand whether the one-time chair stand test (CS-1) is useful for predicting the severity of coronavirus disease (COVID-19) in 101 patients admitted to the hospital with acute respiratory failure. Methods T
Externí odkaz:
https://doaj.org/article/f9a411060526449cb54699b86822dad3
The notion of flat minima has played a key role in the generalization studies of deep learning models. However, existing definitions of the flatness are known to be sensitive to the rescaling of parameters. The issue suggests that the previous defini
Externí odkaz:
http://arxiv.org/abs/1901.04653
Autor:
Xiaozhu Wei, Shohei Kumagai, Tatsuyuki Makita, Kotaro Tsuzuku, Akifumi Yamamura, Mari Sasaki, Shun Watanabe, Jun Takeya
Publikováno v:
Communications Materials, Vol 4, Iss 1, Pp 1-9 (2023)
Solution-processable organic thin-film transistors are needed for device applications. Here, solution-processed organic semiconductors and amorphous metal oxide semiconductors are integrated into a transistor, with five-stage complementary ring oscil
Externí odkaz:
https://doaj.org/article/81d5953405104db5b8c0298a1f94a9b9
Autor:
Tsuzuku, Yusuke, Sato, Issei
Data-agnostic quasi-imperceptible perturbations on inputs are known to degrade recognition accuracy of deep convolutional networks severely. This phenomenon is considered to be a potential security issue. Moreover, some results on statistical general
Externí odkaz:
http://arxiv.org/abs/1809.04098
Due to the substantial computational cost, training state-of-the-art deep neural networks for large-scale datasets often requires distributed training using multiple computation workers. However, by nature, workers need to frequently communicate grad
Externí odkaz:
http://arxiv.org/abs/1802.06058
High sensitivity of neural networks against malicious perturbations on inputs causes security concerns. To take a steady step towards robust classifiers, we aim to create neural network models provably defended from perturbations. Prior certification
Externí odkaz:
http://arxiv.org/abs/1802.04034