CASIF Taipei 2023 (CAS Emerging and Selected Topics for Industry Forum - An event by IEEE CASS Taipei Chapter)

#cas #casif #ic #ai #mediatek #tsmc
Share

The development of AI computing has reached a critical inflection point. Large-language models (LLMs) attracted tremendous attention recently, and demonstrated their potential for generating text, images, videos, software programs, and other contents. However, LLMs significantly increase the model parameters size in neural network, and require huge computation for AI model training and inference. For example, The number of parameters of the neural network has grown from 60 million in Alexnet(2012) to 175 billion in OpenAI GPT-3 (2020). Within seven years, the number of parameters has increased by 2917 times , and the computing requirements for training neural networks have increased by 300,000 times. The AI development impacts the IC industry, because the increasing speed of computation requirement far exceeds the growth rate of Moore's Law of semiconductors.

This forum will discuss the current IC design challenge in AI era. It’s more difficult to continuously shrink the semiconductor process to meet the target on performance, power, area, and cost. Huge computing requirement led to high performance, bandwidth, and energy efficiency. Complex IC design also increase the design cost and the demand on highly skilled talents.

On the other hand, AI can enable new functions in IC products, while AI also has potential to be crucial tools to increase the efficiency for IC industry, such as EDA, floor planning, verification, circuit optimization, code generation, etc. It opens new opportunities to revolutionize the IC industry. The forum will also discuss how to leverage AI to improve IC industry.

In particular, the forum will benefit graduate students and young professors, because they can learn about these technical issues through knowledge sharing from experienced experts from industry, academia, and research. This will help young students/professors explore their future career and research paths.

Planned Format of the Forum

This forum will be a half-day conference to include keynote, technical talk and panel discussion. It will bring experts together from leading industry companies, such as TSMC and MediaTek, important research institutes, such as ITRI and TSRI, and top universities, such as National Taiwan University, National Tsing Hua University and National Yang Ming Chiao Tung University. This forum will focus on IC design challenge in AI era, and open opportunities for technical discussion, knowledge sharing, networking and collaboration. The Forum welcome anyone with an interest in this field, including designers, researchers, and engineers from industry, research institute and academia. Especially, we encourage graduate students and young professors to participate this.

It is a hybrid event. The participants are welcomed to attend this event in physical, and video conference casting also provided for registered persons who cannot attend in physical.

To encourage more participants to attend the forum in physical, this event will be co-located with Taiwan Innotech Expo (TIE), which is a huge expo which attracts attendees from IC industries and academics. Last year, TIE 2022 has attracted 18,917 domestic and international visitors. The organizing committee expects co-location with Taiwan Innotech Expo will encourage participants to attend this industrial forum in

Besides, we package three activities: (1) IEEE CASIF Taipei (2) RISC-V Taipei Day (3) YSF Forum (Youth in Semiconductor Technology Forum) in a series with one theme: “IC Tech in the AI Era” and promote three activities together. We hope that will attract more participants, especially young graduate students and young professors.



  Date and Time

  Location

  Hosts

  Registration



  • Date: 12 Oct 2023
  • Time: 09:30 AM to 01:30 PM
  • All times are (UTC+08:00) Taipei
  • Add_To_Calendar_icon Add Event to Calendar
If you are not a robot, please complete the ReCAPTCHA to display virtual attendance info.
  • No. 1, Section 5, Xinyi Road
  • Taipei City, T'ai-pei
  • Taiwan 11049
  • Building: Taipei International Convention Center (TICC)
  • Room Number: 201
  • Click here for Map

  • Contact Event Host
  • IEEE CASS Taipei Chapter 

    Chair :  Bor-Sung Liang  (梁伯嵩博士)  bs.liang@mediatek.com

    Assistant :  Avila Lan (藍彩瑄小姐)  avilalan@ntu.edu.tw

  • Co-sponsored by IEEE Taipei Section, IEEE CASS Tainan Chapter, MediaTek Advanced Research Center, Taiwan Semiconductor Industry Association (TSIA) , Taiwan IC Industry & Academia Research Alliance (TIARA) , RISC-V Taiwan Alliance, AI on Chip Taiwan Alliance (AITA)
  • Survey: Fill out the survey






Agenda

9:00 am – 9:30am / Registration

9:30 am – 9:35 am / Opening 

  • Bor-Sung Liang, Senior Director, Strategic Technology Exploration Platform, MediaTek. Chair, IEEE CAS Taipei Chapter

9:35 am – 9:45 am / Remarks and Group Photo

9:45 am – 10:00 am / Introduction of IEEE CAS Society

  • Kea-Tiong (Samuel) Tang, Professor, Electrical Engineering, National Tsing Hua University. Vice President, Regional Activities and Membership (Executive Committee), IEEE CASS, Region 10

10:00 am – 10:30 am / Keynote

  • Meng-Fan (Marvin) Chang, IEEE Fellow, Director, Corporate Research, TSMC

10:30 am – 11:00 am / Technical talk

  • Allen Lu, Senior Director, Computing and AI Technology Group, MediaTek

11:00 am – 12:00 noon / Panel discussion

  • Shih-Chieh Chang, (Moderator), General Director of Electronic and Optoelectronic System Research Laboratories, Industrial Technology Research Institute (ITRI)
  • Chen-Yi Lee, Senior Vice President, National Yang Ming Chiao Tung University (NYCU)
  • Meng-Fan (Marvin) Chang, IEEE Fellow, Director, TSMC
  • Allen Lu, Senior Director, MediaTek
  • Frankwell Lin, Chairman, Andes Technology
  • Tuo-Hung (Alex) Hou, Director General, Taiwan Semiconductor Research Institute (TSRI)

12:00 noon – 13:30 Lunch



  Media

IEEE_CASIF_Taipei_2023__ver._0729_ 1.01 MiB
2023 CASIF Taipei Report (IEEE CASS Taipei) Agenda and Photos 3.26 MiB