跳到主要內容區

[公告]109學年度上學期專題指導教授同意書下載與相關說明

各位同學好,

103學年度第1學期第1次資工系課程委員會決議,103年度起取消學生開始找指導教授及老師開始收學生之時間點規定,

僅規定學生最後繳交專題指導教授申請表之時間。

106學年度起,本系學生可尋找電資院專任師資(包含資工系與電機系)擔任專題指導教授,唯學分計算、成績評分、報告方式等需依照本系規定辦理。

此外,每名教師招收專題生資工系同年級最多5名。

請同學們與老師商訂題目並線上填寫指導教授同意書,印出由老師簽名後,於2021/01/15(五)前交給班代統一收齊,送回系辦大學部事務承辦人處。

由指導教授簽名之同意書紙本務必繳回系辦,未繳交者視同無尋找指導教授,屆時無法登打專題成績。

指導教授同意書填寫連結:

http://web.cs.nthu.edu.tw/p/423-1174-2989.php

若有任何問題,歡迎與系辦公室之大學部事務承辦人聯繫,謝謝您。

系辦公室敬啟

2020.09.14.

Dear students,

For CS 3901 System Integration Implementation, you may start looking for an adviser once this announcement has been made.

A student may work on a project supervised by any EECS professor of our college subject to all requirements under the rules of the CS department.

Every EECS professor may supervise up to 5 CS students admitted to the university in the same year.

Please complete the application form online (
http://web.cs.nthu.edu.tw/p/423-1174-2989.php),

print it out and ask your project adviser to sign it. Give the signed form to your class representative no later than January 4, 2021.  

If you have any question, please contact the staff for undergraduate affairs in the department office. Thank you.

CS Office

2020.09.14.

 

師長所提供的專題題目與說明資訊,將於09/14(一)開始公告及更新於下列,若無在表列中,請直接洽詢該位師長。

 (表列更新日期:2020/9/14)

 

教授姓名  專題題目  專題說明  是否自辦說明會?  日期/時間  地點  備註 
孫宏民 資訊安全,量化交易 ◆ 資訊安全:
目前實驗室的研究資源為App安全檢測和無線網路安全攻防相關,但也歡迎同學對其他有興趣的資安領域來做探討

◆ 量化交易: 目前主要是以台指期做相關研究,並進行程式交易策略之撰寫

◆注意事項: 此二領域(尤其資安)需大量相關背景知識,因此實驗室會安排一些課程,建議有興趣的同學先確定能掌握這些知識後再決定是否加入。之後也會要求參與一些資安相關競賽(CTF, 金盾獎等等)
否(請直接按送出) /    
李政崑 Software Project 1. Topic 1: Design a RNN APP (AI application APP)

2. Topic 2: LLVM Compiler Optimization
否(請直接按送出) /   e-mail: jklee@cs.nthu.edu.tw or
graduate student: cclin@pllab.cs.nthu.edu.tw
王廷基 1. Design and implementation of efficient/robust algorithms for circuit placement or routing.
2. Adversarial attack or defense for neural networks.
Students who have strong interests in programming, data structures, algorithms, and machine learning are welcome to email me (tcwang@cs.nthu.edu.tw) for an appointment. 否(請直接按送出) /    
林華君 醫療資訊處理   否(請直接按送出) /    
黃稚存 深度學習與深度神經網路 1. 仿人機器手臂深度學習控制及機器視覺
2. 人工智慧運動分析 (如AI棒球投打分析)

可深入認識人工智慧、深度學習各種關鍵的技術,也能瞭解它目前的限制。歡迎想動手研究、解決真正問題的同學加入。
否(請直接按送出) /    
麥偉基 CAD for Quantum Computing and/or Superconducting Logic Circuits The project student(s) is expected to explore and implement reinforcement learning approaches to solve one of the new and challenging problems below:
- Mapping Quantum Circuits to quantum computer architectures such as IBM QX
- Placement of rapid single flux quantum (RSFQ) circuits considering wire-length matching
否(請直接按送出) /    
王炳豐 程式設計與演算法訓練 資格: 願意花時間增強程式與解決問題能力, 學習如何寫出聰明漂亮程式的同學, 程度不拘, 但需要願意花時間投入

說明
:
1.
所有人對資工系畢業生最主要的期待, 就是能寫出聰明漂亮的程式,
這也是資工系畢業生最主要的競爭力來源
2. 專題將以歷年 ACM 世界賽題目為教材, 透過討論讓同學有能力解決每一個世界賽中的難題,
學會寫出聰明漂亮的程式
3. 訓練的重點在於演算法的設計, 證明, 與分析;
不是單純的實作訓練

聯絡
: 請或送 email 和我聯絡 bfwang@cs.nthu.edu.tw
否(請直接按送出) /    
金仲達 多重時序機械手臂動作之強化學習與應用 使用強化學習訓練機械手臂進行須由多段子動作串接方能完成之任務。 否(請直接按送出) /    
徐正炘 XR Cloud Gaming, Point Cloud Compression and Streaming, Smart Home, Federated Learning, 6DoF XR streaming VR cloud gaming:
As cloud gaming and Virtual Reality (VR) games catch more and more people's eyes in the game industry, game developers engage in these fields to publish better merchandise. That cloud gaming possesses the merit of lifting heavy computation loads from client device to server solves the high resource consumption on VR games for regular clients. Consequently, it is important to know where is the bottleneck of the platform and how can the VR cloud gaming platform be improved in the future. Based on an open-sourced remote wireless VR rendering project, Air Light Virtual Reality (ALVR), we build a cloud VR gaming testbed. We aim to add new features, such as QUIC, adaptive bitrate, and super-resolution, so as to optimize the performance and provide a better user experience.

Point Cloud Compression and Streaming:
Point Cloud is the ideal data representation for various applications, e.g., 3D scene reconstruction, 6DoF XR. However, streaming raw point cloud video needs a lot of bandwidth, which is far higher than the available bandwidth today. In this project, we study the point cloud compression and streaming optimization to address the aforementioned problem.

Smart Home:
We collaborate with Prof. Shervin from the University of Ottawa, Canada on this project. This project focuses on food activity recognition to know when, where, and what food intake activities people are doing in a smart home. For the current state, we first leverage millimeter-wave radar as our data source. Millimeter-wave has a higher frequency so that it’s more precious, robust, and able to preserve more privacy compared with RGB cameras. The size of it is quite small to be concealed unobtrusively and it is capable of adapting the dynamic environment. We collect, train, and recognize point clouds that captured from millimeter wave sensors. Our goal is to recognize food activities in human daily lives with minimal interference.

Federated Learning
We're currently learning the federated learning concept by hosting a presentation meeting weekly. Federated learning runs the training process at multiple end devices having the training data and collects the weights to a higher layer to update the global model. By doing so, federated learning ensure privacy (the privacy-sensitive raw data won't be transmitted to other devices). During the meeting, we'll present related works to lab mates, showing the progress of our implementation, and discuss the next step. The final goal is to design and implement a privacy-aware intelligent platform at the edge of our campus.

6DoF XR Streaming:
Virtual reality (VR) and Augmented reality (AR) technology are very popular in recent years. Traditionally, VR and AR only support three Degree of Freedom (3DoF), which allow view changing according to the viewer's orientation.  6DoF is the new interaction way, which changes view according to the viewer's position and orientation. It provides a more immersive experience to the user. In this project, we study the system and network supports of 6DoF XR streaming.
是(請續填以下資訊) 20200917/17:40 台達館 613 當天提供餐飲
蔡仁松 資安晶片及系統設計 鑑於資安成為國安問題,參加學生將參與由博班學生帶領的研究資安晶片及系統設計。 否(請直接按送出) /    
瀏覽數:
登入成功