課程資訊
課程名稱
 專題討論
Seminar 
開課學期
 105-2 
授課對象
 電子工程學研究所  
授課教師
 江介宏 
課號
 EEE7001 
課程識別碼
 943 M0020 
班次
 03 
學分
 0.0 
全/半年
 半年 
必/選修
 必修 
上課時間
 星期一8(15:30~16:20) 
上課地點
  
備註
 EDA組碩士班在學期間每學期必修，上課地點：博理112
限本系所學生(含輔系、雙修生)
總人數上限：100人 
Ceiba 課程網頁
 http://ceiba.ntu.edu.tw/1052EEE7001_03 
課程簡介影片
 
核心能力關聯
 核心能力與課程規劃關聯圖
課程大綱
為確保您我的權利,請尊重智慧財產權及不得非法影印
課程概述

EDA Seminar 

課程目標
 Seminar 
課程要求
 待補 
預期每週課後學習時數
  
Office Hours
  
指定閱讀
 待補 
參考書目
 待補 
評量方式
(僅供參考)
    
課程進度
週次
日期
單元主題
第1週
 2/20  Introduction 
第2週
 2/27  -- (Holiday) 
第3週
 3/06  陳有棟協理 奇景光電
講題：談工程師信心的建立
講員簡介：成功大學電機系碩士、學士，建國中學 / 奇景光電協理、處長 / 凌陽科科技處長 / 我想科技 (鈺創科技)經理、處長 / 聯陽半導體資深工程師
2013年加入奇景光電，專長為類比/數位混合信號處理電路及系統設計，高速介面電路設計，射頻及寬頻電路設計，目前負責設計工程中心，主要負責影像處理IP開發、高速影像傳輸介面IP開發、PI/SI/EMI 及ESD 開發設計及晶片實現等Physical Design 工作。
 
第4週
 3/13  Prof. Iris Hui-Ru Jiang, Department of Electronics Engineering, National Chiao Tung University
江蕙如教授, 國立交通大學電子工程學系

Title:
I’m Starting to See a Pattern Here…

Abstract:
Lithography layout decomposition and layout pattern classification are critical and indispensable processes for today's high performance chip design and manufacturing. In this talk, we present two fast and general algorithms on these topics. We also demonstrate that in addition to solution quality, nice algorithms rely on good data structures for achieving high efficiency.
As the technology node advances, more complex coloring rules are introduced. In the first part, for capturing the essence of multiple patterning layout decomposition (MPLD) with complex coloring rules, we model this problem as an exact cover problem and propose a fast and exact MPLD framework based on augmented dancing links. We also develop special treatments based on the properties of MPLD to expedite the solving. Our method is flexible (considering basic and complex coloring rules, and density balancing) and general (handling arbitrary k-patterning). Our results show that our approach achieved the least number of conflicts and stitches among state-of-the-art works, attained superior efficiency and scalability, and handled complex coloring rules and density balancing well.
Layout pattern classification underlies a variety of design for manufacturability applications. In the second part, we present a fast and general layout pattern classification algorithm. Our clip representation can handle not only rigid area match or edge displacement constraints but also variant edge tolerances and don’t care regions. For achieving small cluster count, our clip clustering follows the natural grouping structure of layout clips. For further enhancing efficiency, we propose two-stage clustering including identical clip merging and Markov clustering with feasibility recovery. Our results show that our algorithm outperforms the reference solution and all winning teams of 2016 CAD contest at ICCAD, delivering the smallest cluster count, fastest runtime, and 100% validity.

Bio:
Iris Hui-Ru Jiang received the B.S. and Ph.D. degrees in electronics engineering from National Chiao Tung University (NCTU), Hsinchu, Taiwan, in 1995 and 2002, respectively. She is currently a Professor with the Department of Electronics Engineering, NCTU and has been a visiting scholar of IBM Austin Research Laboratory from 2013 to 2014. Her current research interests include interaction between logic design and physical synthesis, timing analysis and optimization, physical design optimization, design for manufacturability, and data analytics based design automation.
Dr. Jiang received Best Paper Award Nomination from DAC 2016 and ISPD 2013 and Best in-track Paper from ICCAD 2014. She and her students was the recipient of the First Place Award at the CAD Contest at ICCAD in 2012. Her group has received Awards at the TAU Timing Analysis Contests (5 years in a row since 2013). She was the recipient of the 2011 Chinese Institute of Electrical Engineering Outstanding Young Electrical Engineer Award. She is currently the chair of IEEE CEDA Design Automation Technical Committee (DATC), and organized CAD contest at ICCAD from 2012 to 2014 and CADathlon@ICCAD since 2016. She is currently an associate editor of IEEE TCAD and has served on technical program committees of major EDA conferences, including DAC, ICCAD, ISPD, ASP-DAC, SLIP, and IWLS. 
第5週
 3/20  Dr. Hermes Lin, Synopsys

Titile:
On verification practices in EDA industry and its trends

Abstract:
As the complexity of chip design is growing exponentially in past few decades, how to verify a chip design to ensure chip work as expected is an ongoing challenge task for both chip designer and EDA tool provider.
The challenges come from not only chip complexity but also the pressure of shorter time to market.
In this talk, Dr. Lin will share some practices on solving this verification challenge and also the evolving trend from his observation in industry.

Short Bio:
Hermes Lin was born on June, 1972 in Kaoshiung City, Taiwan. He received Ph. D. degree in electronics engineering from National Chao Tung University, Hsinchu, Taiwan in 2000.
During his Ph. D. program, his research interests include hardware description language (HDL) synthesis, graph theory and high-level/logic synthesis.
Since 2001, he joined industrial and focused on verification area. He joined Novas Software, Inc., California during 2000~2003 and SpringSoft Inc, Hsincu, during 2004~2012 and worked on leading edge debug platform.
He joined Synopsys since year 2012 and has joined the development of “verification continuum” till now. He now is R&D director of Synopsys Verification Group.
 
第6週
 3/27  郁方教授 國立政治大學 資管所

Title: String Analysis

Abstract: The goal of string analysis techniques is to determine the
set of values that string expressions can take during program
execution. Like many other program analysis problems, it
is not possible to precisely determine the set of string values
that can reach a program point. However, one can compute
over or under-approximations of possible string values. If the
approximations are precise enough, they may enable us to
demonstrate existence or non-existence of bugs in string ma-
nipulating code. String analysis has been an active research
area in the last decade, resulting in a wide variety of string
analysis techniques. In this talk, we
will discuss automated string analysis techniques, focusing
particularly on automata-based static string analysis.
The same talk was previously given in the tutorial of ACM SIGSOFT FSE 2014.

Bio:
Dr. Fang Yu is an Associate Professor in the Department of Management Information Systems at National Chengchi University. He was an Assistant Professor in 2010-2014. He received his PhD degree and the 2010 outstanding dissertation award in Computer Science from University of California, Santa Barbara. His research interests broadly span software security, verification, formal methods, automata theory, and membrane and neural computing. He published papers in ICSE, FSE, ISSTA, ASE, TACAS, CAV, WWW, DSN, etc. and TSE and FMSD.
He has served the program committees of ACM LCTES, IEEE CloudCom, IEEE SCC, IEEE BigData, ATVA, etc. and co-chaired Infinity 2011. He served the executive editor of International Journal of Information and Computer Security in 2014-2015. He was the associate research scientist of the DARPA project during the visit to VLab@UCSB in 2016. 
第7週
 4/03  -- (Holiday) 
第8週
 4/10  At his Stanford University commencement speech, Steve Jobs, CEO and co-founder of Apple and Pixar, urges us to pursue our dreams and see the opportunities in life's setbacks — including death itself.

https://www.ted.com/talks/steve_jobs_how_to_live_before_you_die 
第9週
 4/17  -- (Midterm Week) 
第10週
 4/24  Entrepreneur Elon Musk is a man with many plans. The founder of PayPal, Tesla Motors and SpaceX sits down with TED curator Chris Anderson to share details about his visionary projects, which include a mass-marketed electric car, a solar energy leasing company and a fully reusable rocket.

https://www.ted.com/talks/elon_musk_the_mind_behind_tesla_spacex_solarcity
 
第11週
 5/01  JENNIFER DOUDNA: GENOME ENGINEERING WITH CRISPR-CAS9: BIRTH OF A BREAKTHROUGH TECHNOLOGY
Jennifer Doudna tells the story of how studying the way bacteria fight viral infection turned into a genomic engineering technology that has transformed molecular biology research. In 2013, Doudna and her colleagues developed the CRISPR-Cas9 gene expression system that, when introduced into animal cells, makes site-specific changes to intact genomes. CRISPR-Cas9 is more precise, more efficient, and less expensive than other genome editing tools and, as a result, has facilitated a wide range of studies that were previously unachievable.

https://www.ibiology.org/ibiomagazine/jennifer-doudna-genome-engineering-with-crispr-cas9-birth-of-a-breakthrough-technology.html
 
第12週
 5/08  揚智科技 吳文宏處長 
第13週
 5/15  TSMC Speaker 
第14週
 5/22  凌嘉鴻博士 中央研究院生化所 臺灣大學生化科學研究所

Title:
Overcoming challenges in CRISPR/Cas9 gene editing

Abstract:
The CRISPR/Cas9 system is a robust genome editing technology that works in human cells, animals, plants and microbes based on the RNA-programmed DNA cleaving activity of the Cas9 enzyme. This presentation provides a quick overview of the fascinating journey through which the studies of CRISPR microbial immune system rapidly revolutionizes the research in genome and cell biology. We will go over the current development of the CRISPR/Cas9 technology, and the strength and weakness of this system. Through real case studies, we will understand the unsolved problems in CRISPR/Cas9-mediated genome engineering and discuss the challenges to overcome to realize its therapeutic potential.
 
第15週
 5/29  -- (Holiday) 
第16週
 6/05  陳東傑博士 Dr. Tung-Chieh Chen
至達科技 Maxeda Technology

Title: EDA工作經驗分享

Abstract:
念了EDA組不知道將來要做什麼工作？
本次演講，陳博士將和大家分享在EDA領域工作的經驗。
內容包含EDA職場職務介紹，職場工作發展，
職場程式設計和學校程式設計的差異等。


Bio:
陳東傑
至達科技執行長

經歷
2017 臺大電子所 電子菁英特殊貢獻獎
2008 臺灣大學電子所博士
2007 國際積體電路馬拉松程式競賽（ACM CADathlon at ICCAD）冠軍
2006 ACM ISPD & IEEE CEDA 世界積體電路擺置軟體開發競賽季軍
教育部大學校院積體電路電腦輔助設計軟體競賽多次冠軍
頂尖國際會議（ICCAD，ASPDAC，ISPD）技術委員
思源科技與新思科技研發經理
美國專利11件，IEEE/ACM 電腦輔助設計自動化相關期刊與會議論文32篇，專書4本

陳東傑博士目前為至達科技執行長。至達科技為一家電子設計自動化公司，針對先進製程的超大型積體電路設計提供高速最佳化、最有效及最高規格的解決方案。陳博士曾任職於全台最大且唯一上市的電子設計自動化公司思源科技，將當時學術界最強大的自動電路擺置系統NTUplace3 的核心移植至思源科技的主力產品 Laker 客製化數位擺置器，並開發 Laker L4 類比電路版圖原型系統，使得 Laker 能力大增，深受客戶好評。此舉更影響了全世界最大的電子設計自動化公司新思科技在 2012 年中以 122 億新台幣高價併購了思源科技。此外，陳博士兩度共同指導台灣大學電子工程學研究所的研究團隊獲得全球最大型的 ACM/IEEE 積體電路軟體研發競賽第一名（此競賽被 EE Times 技術編輯喻為“地表最優秀工程人才的競技場”）。 
第17週
 6/12  師生座談