:::

Activity

2017 International Symposia on VLSI

The ITRI-sponsored International Symposium on VLSI Technology, Systems and Applications (VLSI-TSA) and International Symposium on VLSI Design, Automation & Test (VLSI-DAT) kicked off on April 25. This year’s discussions focused on the state of development and future challenges in IoT-enabling technologies, wearable devices, unmanned aerial vehicles (UAV), virtual reality/augmented reality, robotics, and artificial intelligence. Representatives from top tier schools and companies, including the University of California, Northwestern University, National Taiwan University, National Chiao Tung University, ARM Holdings, and HTC, shared the latest developments in semiconductor components and manufacturing processes as well as trends in chip design and systems integration applications.

Dr. Chih-I Wu, VLSI-TSA Co-chair and General Director of ITRI’s Electronic and Optoelectronic System Research Laboratories, stated that rapid changes in technology are pushing Taiwan’s industries to engage in the development of smart systems and multi-disciplinary integration. The combination of capability in AI software and data analytics and existing strengths in ICT, optoelectronics, and biotechnology will enable Taiwan to provide comprehensive solutions, thereby identifying new positions and opportunities in the marketplace, he said.

Dr. Tzi-Cker Chiueh, VLSI-DAT Executive Committee Member and General Director of ITRI’s Information and Communications Research Laboratories, expects the focus of AI development in the next several years will be on Deep Neural Network (DNN) algorithms, a new area for machine learning. DNN is a mathematical model that mimics the biological nervous system and allows developed programs to learn on their own. He suggested that Taiwan can choose to develop DNN programs in areas such as legal document analysis, patent application analysis, and medical record analysis. Consequently, Taiwan will be able to play a key role in AI business applications.

At a joint plenary session on the first day of the VLSI symposium, Prof. Subramanian S. Iyer from the University of California, Los Angeles delivered an address on “Heterogeneous SoCs.” Prof. Iyer said that the size of SoCs has become increasingly smaller and that the development of a heterogeneous parallel interface is reducing the size of PCBs. These methods promise to reduce the NRE by 10-20 times and shorten the time to market for products.

Meanwhile, Prof. Ren C. Luo, the Irving T. Ho Chair and Distinguished Professor in the Department of Electrical Engineering at National Taiwan University, spoke on the topic of “World Megatrends of Intelligent Robotics and AI.” He stated that as society’s need for robots is becoming pronounced, AI has become an important core technology with broad applications including smart robots and speech recognition. The global AI market is estimated to exhibit a compound growth rate of 23%, reaching a scale of US$500 million in 2020. As the market for semiconductor killer applications for PCs, mobile phones, and TVs wanes, the megatrend now is toward smart city, smart factory, and smart home applications that require high-performance computing, ultra-low power sensors, and MCUs.