Search Results - "Sakai, Yasufumi"

A 1.95 GHz Fully Integrated Envelope Elimination and Restoration CMOS Power Amplifier Using Timing Alignment Technique for WCDMA and LTE by Oishi, Kazuaki, Yoshida, Eiji, Sakai, Yasufumi, Takauchi, Hideki, Kawano, Yoichi, Shirai, Noriaki, Kano, Hideki, Kudo, Masahiro, Murakami, Tomotoshi, Tamura, Tetsuro, Kawai, Shigeaki, Suto, Kazuo, Yamazaki, Hiroshi, Mori, Toshihiko

“…A fully integrated envelope elimination and restoration (EER) CMOS power amplifier (PA) has been developed for WCDMA and LTE handsets. EER is a supply…”
Get full text

Bioaccumulation and Toxic Potencies of Polychlorinated Biphenyls and Polycyclic Aromatic Hydrocarbons in Tidal Flat and Coastal Ecosystems of the Ariake Sea, Japan by Nakata, Haruhiko, Sakai, Yasufumi, Miyawaki, Takashi, Takemura, Akira

“…Sediment and marine biota comprising several species of tidal flat and coastal organisms were analyzed for polychlorinated biphenyls (PCBs) including non- and…”
Get full text

S-DFP: shifted dynamic fixed point for quantized deep neural network training by Sakai, Yasufumi, Tamiya, Yutaka

“…Recent advances in deep neural networks have achieved higher accuracy with more complex models. Nevertheless, they require much longer training time. To reduce…”
Get full text

Greedy search algorithm for partial quantization of convolutional neural networks inspired by submodular optimization by Tsuji, Satoki, Yamada, Fuyuka, Kawaguchi, Hiroshi, Inoue, Atsuki, Sakai, Yasufumi

Get full text

Automatic Pruning Rate Derivation for Structured Pruning of Deep Neural Networks by Sakai, Yasufumi, Iwakawa, Akinori, Tabaru, Tsuguchika, Inoue, Atsuki, Kawaguchi, Hiroshi

“…To compress the neural network model, structured pruning has been proposed. However, finding a proper pruning rate to suppress the accuracy degradation of…”
Get full text

High-throughput computational screening of doped transition metal oxides as catalysts for nitrogen reduction by Höskuldsson, Árni Björn, Dang, Thang, Sakai, Yasufumi, Ishikawa, Atsushi, Skúlason, Egill

Get full text

Dropout and DropConnect for Reliable Neuromorphic Inference Under Communication Constraints in Network Connectivity by Sakai, Yasufumi, Pedroni, Bruno U., Joshi, Siddharth, Tanabe, Satoshi, Akinin, Abraham, Cauwenberghs, Gert

“…Dropout and DropConnect are known as effective methods to improve on the generalization performance of neural networks, by either dropping states of neural…”
Get full text

Quantizaiton for Deep Neural Network Training with 8-bit Dynamic Fixed Point by Sakai, Yasufumi

“…Recent advances in deep neural networks have achieved higher accuracy with more complex models. Nevertheless, they require much longer training time. To reduce…”
Get full text

Envelope tracking CMOS power amplifier with high-speed CMOS envelope amplifier for mobile handsets by Yoshida, Eiji, Sakai, Yasufumi, Oishi, Kazuaki, Yamazaki, Hiroshi, Mori, Toshihiko, Yamaura, Shinji, Suto, Kazuo, Tanaka, Tetsu

“…A high-efficiency CMOS power amplifier (PA) based on envelope tracking (ET) has been reported for a wideband code division multiple access (W-CDMA) and long…”
Get full text

Active Learning for Graph Neural Networks Training in Catalyst Energy Prediction by Sakai, Yasufumi, Matsumura, Naoki, Inoue, Atsuki, Kawaguchi, Hiroshi, Thang, Dang, Ishikawa, Atsushi, Hoskuldsson, Arni Bjorn, Skulason, Egill

“…The electrification of major industrial processes constitutes an important step in reducing global carbon emissions. Thus, the identification of materials able…”
Get full text

3.5 A 56Gb/s NRZ-electrical 247mW/lane serial-link transceiver in 28nm CMOS by Shibasaki, Takayuki, Danjo, Takumi, Ogata, Yuuki, Sakai, Yasufumi, Miyaoka, Hiroki, Terasawa, Futoshi, Kudo, Masahiro, Kano, Hideki, Matsuda, Atsushi, Kawai, Shigeaki, Arai, Tomoyuki, Higashi, Hirohito, Naka, Naoaki, Yamaguchi, Hisakatsu, Mori, Toshihiko, Koyanagi, Yoichi, Tamura, Hirotaka

“…With the rapid growth of data traffic in data centers, data rates over 50Gb/s/signal (e.g., OIF-CEI-56G-VSR) will eventually be required in wireline…”
Get full text

GPQ: Greedy Partial Quantization of Convolutional Neural Networks Inspired by Submodular Optimization by Tsuji, Satoki, Yamada, Fuyuka, Kawaguchi, Hiroshi, Inoue, Atsuki, Sakai, Yasufumi

“…Recent work has revealed that the effects of neural network quantization on inference accuracy are different for each layer. Therefore, partial quantization…”
Get full text

Efficient and Large Scale Pre-training Techniques for Japanese Natural Language Processing by Kasagi, Akihiko, Asaoka, Masahiro, Tabuchi, Akihiro, Oyama, Yosuke, Honda, Takumi, Sakai, Yasufumi, Dang, Thang, Tabaru, Tsuguchika

“…Pre-training in natural language processing greatly affects the accuracy of downstream tasks. However, pre-training is a bottleneck in the AI system…”
Get full text

3.2 A 1.95GHz fully integrated envelope elimination and restoration CMOS power amplifier with envelope/phase generator and timing aligner for WCDMA and LTE by Oishi, Kazuaki, Yoshida, Eiji, Sakai, Yasufumi, Takauchi, Hideki, Kawano, Yoichi, Shirai, Noriaki, Kano, Hideki, Kudo, Masahiro, Murakami, Tomotoshi, Tamura, Tetsuro, Kawai, Shigeaki, Yamaura, Shinji, Suto, Kazuo, Yamazaki, Hiroshi, Mori, Toshihiko

“…In recent years, the demand for low cost and system-on-a-chip for mobile terminals has led to the development of a highly-integrated, low-distortion, and…”
Get full text

DropOut and DropConnect for Reliable Neuromorphic Inference under Energy and Bandwidth Constraints in Network Connectivity by Sakai, Yasufumi, Pedroni, Bruno U., Joshi, Siddharth, Akinin, Abraham, Cauwenberghs, Gert

“…DropOut and DropConnect are known as effective methods to improve on the generalization performance of neural networks, by either dropping states of neural…”
Get full text

A 28.3 Gb/s 7.3 pJ/bit 35 dB backplane transceiver with eye sampling phase adaptation in 28 nm CMOS by Miyaoka, Hiroki, Terasawa, Futoshi, Kudo, Masahiro, Kano, Hideki, Matsuda, Atsushi, Shirai, Noriaki, Kawai, Shigeaki, Shibasaki, Takayuki, Danjo, Takumi, Ogata, Yuuki, Sakai, Yasufumi, Yamaguchi, Hisakatsu, Mori, Toshihiko, Koyanagi, Yoichi, Tamura, Hirotaka, Ide, Yutaka, Terashima, Kazuhiro, Higashi, Hirohito, Higuchi, Tomokazu, Naka, Naoaki

“…28.3 Gb/s transceiver with 35 dB channel loss equalization is presented. The transmitter deploys 3-tap feed forward equalizer (FFE). The driver employs the…”
Get full text

Notes on the occurrence of pathogenic races of xanthomonas oryzae pv. oryzae found in Hiroshima prefecture [Japan] in 1999 by Oku, T. (Hiroshima Prefectural Univ., Shobara (Japan)), Sato, K, Kurao, M, Matsuura, S, Sakai, Y, Tsuchiya, T

Get full text