|
AI研究方向及摘要:
主要研究方向是在於低功率(low-power)、低延遲(low-latency)的學習晶片:使用仿神經(neuromorphic)的架構,打破傳統馮紐曼(Von-Neumann)架構的限制,並使用突波網路(spike-based),達到極高效率的AI晶片系統設計。這需要軟體與硬體共同結合的開發,同時需要模型壓縮與電路設計相輔相成的搭配。
AI相關發表論文 (詳見 http://140.114.14.182/drupal7_qs/Pub):
#Shih-Wen Chiu, #Jen-Huo Wang, #Kwuang-Han Chang, #Ting-Hau Chang, #Chia-Min Wang, #Chia-Lin Chang, Chen-Ting Tang, Chien-Fu Chen, Chung-Hung Shih, Han-Wen Kuo, Li-Chun Wang, Hsin Chen, Member, IEEE, Chih-Cheng Hsieh, Meng-Fan Chang, Yi-Wen Liu, Tsan-Jieh Chen, Chia-Hsiang Yang, Herming Chiueh, Juyo-Min Shyu, and Kea-Tiong Tang*, “A Fully Integrated Nose-on-a-Chip for Rapid Diagnosis of Ventilator-Associated Pneumonia”, IEEE Transaction on Biomedical Circuits and Systems, vol. 8(6), pp. 765-778, 2014. (5-Year IF = 3.146, Ranking: 27/243
Shih-Wen Chiu, Hsiang-Chiu Wu, Ting-I Chou, Hsin Chen, and Kea-Tiong Tang*, “A Miniature Electronic Nose System Based on a MWNT-Polymer Microsensor Array and a Low-Power Signal Processing Chip”, Analytical & Bioanalytical Chemistry, pp. 1-10, 2014/01/03, 2014. (5-Year IF = 3.756, Ranking: 9/75)