```
├── CV/
   ├── cv-eylai.pdf
├── GRE/
   ├── AW/
      ├── GRE 写作-244道Issue题目分析与提纲.pdf
   ├── CS_CS-Adjacent GRE Requirements.xlsx
   ├── Math/
      ├── 巍哥GRE数学170难题3.0.pdf
      ├── 巍哥GRE数学满分宝典.pdf
   ├── RC/
      ├── 2021版GREfo脚阅读400系列-Volume 1-阅读机经篇.pdf
      ├── 新GRE阅读逻辑10套.pdf
      ├── 杨鹏长难句 完整版.pdf
   ├── Vocab/
      ├── GRE镇考机经词6.0【微信公众号：张巍老师GRE】.pdf
      ├── 再要你命3000.pdf
      ├── 新GRE词汇进阶与巩固201407.pdf
├── LICENSE (omitted)
├── My Application Documents/
   ├── Academic_Statement_of_Purpose_UIUC_.pdf
   ├── HKPFS.md (4.2k tokens)
   ├── Personal_Statement_NUS_.pdf
   ├── Personal_Statement_UIUC_.pdf
   ├── Research_Proposal_on_HRL.pdf
   ├── Research_Proposal_on_TAMP.pdf
   ├── Source Code/
      ├── Academic Statement of Purpose(UIUC).zip
      ├── Personal Statement(NUS).zip
      ├── Personal Statement(UIUC).zip
      ├── Research Proposal on HRL.zip
      ├── Research Proposal on TAMP.zip
      ├── recommendation-letter-chen.doc
      ├── recommendation-letter-lu.docx
   ├── recommendation-letter-chen.pdf
   ├── recommendation-letter-lu.pdf
├── Personal Statement/
   ├── ShortPersonalStatement.pdf
   ├── StatementOfPurpose-MIT.pdf
   ├── StatementOfPurpose-UCB.pdf
   ├── UMPersonalStatement.pdf
   ├── personal_statement.pdf
   ├── ps-cal-eylai.pdf
├── Practical Experience/
   ├── 浙大飞跃手册.pdf
├── README.md (1100 tokens)
├── Recommendation Letter/
   ├── THU Letter of Recommendation Template.zip
├── Research Statement/
   ├── A Topological Framework for Visualizing Time-varying Volumetric Datasets.pdf
   ├── Research_Statement_DM.pdf
   ├── research_statement_UIUC_CS.pdf
├── Statement of Purpose/
   ├── UMStatementofPurpose.pdf
   ├── phd_admission_sop.pdf
   ├── sop-mit-eylai.pdf
   ├── statement-of-purpose.pdf
```


## /CV/cv-eylai.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/CV/cv-eylai.pdf

## /GRE/AW/GRE 写作-244道Issue题目分析与提纲.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/AW/GRE 写作-244道Issue题目分析与提纲.pdf

## /GRE/CS_CS-Adjacent GRE Requirements.xlsx

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/CS_CS-Adjacent GRE Requirements.xlsx

## /GRE/Math/巍哥GRE数学170难题3.0.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/Math/巍哥GRE数学170难题3.0.pdf

## /GRE/Math/巍哥GRE数学满分宝典.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/Math/巍哥GRE数学满分宝典.pdf

## /GRE/RC/2021版GREfo脚阅读400系列-Volume 1-阅读机经篇.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/RC/2021版GREfo脚阅读400系列-Volume 1-阅读机经篇.pdf

## /GRE/RC/新GRE阅读逻辑10套.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/RC/新GRE阅读逻辑10套.pdf

## /GRE/RC/杨鹏长难句 完整版.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/RC/杨鹏长难句 完整版.pdf

## /GRE/Vocab/GRE镇考机经词6.0【微信公众号：张巍老师GRE】.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/Vocab/GRE镇考机经词6.0【微信公众号：张巍老师GRE】.pdf

## /GRE/Vocab/再要你命3000.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/Vocab/再要你命3000.pdf

## /GRE/Vocab/新GRE词汇进阶与巩固201407.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/GRE/Vocab/新GRE词汇进阶与巩固201407.pdf

## /My Application Documents/Academic_Statement_of_Purpose_UIUC_.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Academic_Statement_of_Purpose_UIUC_.pdf

## /My Application Documents/HKPFS.md

**Please describe your related experience or participation in research projects**

While my past experiences have been diverse, research has unquestionably been an important composition during my education and career. Back to my time at Imperial College London, participating in laboratory research has been a fantastic complement to my undergraduate studies, strengthening my conceptual understanding of the material taught in class. After I have joined the industry, research work and the ensuing interactions with my colleagues continue to provide me with a toolbox of hard and soft skills that has bonus in my life. Most of my research experiences are intimately related to robotics and automation. In the following paragraphs, it will be shown that how the dots in my life have been connected to bring me where I am today: an enthusiast for robotics.

My passion for robotics arose during my time at university, where I worked as an undergraduate research assistant in Robot Intelligence Lab (Summer 2017), supervised by Dr. Petar Kormushev. This period allowed me to conduct an individual project whose primary goal was to build a three-fingered underactuated dexterous robotic hand for a Baxter robot in the laboratory. The end effector can automatically adapt to any object, with adjustable torque and speed. It was my first time getting exposure to the world of robotics.  I have had the chance to master essential research techniques in robotics, including ROS, forward/inverse kinematics and motion planning. Moreover, I acquired an engineering mindset of modular systems thinking. Having time to learn these techniques under close supervision has improved my efficiency in the lab and has given me confidence in my future research and studies.

In the fall of 2018, I took an industrial placement at 10X research team, Ocado Technology. The team has been working on a project towards the next generation of warehouse robots roaming on 3D racks in the fulfillment center. My main contribution to this project has been the development of linear PMSM (permanent magnet synchronous motor) wheels for the Ocado warehouse robots. I have been fortunate to gain independence in my internship and came up with several innovative ideas. For example, I designed a novel PCB (printed circuit board) rotor that replaces the copper windings for it has been proven to have better acceleration and higher overload current capacity. The outcomes of this project were presented to my peers and advisors at the company-wide internal conference. Having experience demonstrating my work to the audience significantly refined my competence to exchange for innovative ideas with the public. My second project at Ocado Technology was developing simulation software that was used for multi-robot navigation. My contribution was researching on decentralized cooperative multi-agent reinforcement learning algorithms that enhance warehouse efficiency when the number of robots rises dramatically. This project piqued my interest in multi-agent systems and reinforcement learning. I have started thinking about how robots can efficiently corporate with each other without sacrificing safety constraints. Successful completion of this internship provided me with a comprehensive understanding of the world of robotics as a whole: from low-level motors to high-level embodied AI. Moreover, working with Dr. David Sharp, who was a research scientist with a background in both engineering and consultancy, allowed me to view how research work can be transformed into real industrial applications. 

In June 2019, I finished my master's thesis in 'Blind Source Separation Based on Sparsity' supervised by Dr. Wei Dai. Blind source separation (BSS) is a prevalent technique in array processing and data analysis that aims to recover unknown signals from observed mixtures. During this research project, I improved the well-known K-SVD BSS algorithm by introducing a block sparsified dictionary, which clusters the dictionary atoms according to their similarity and those atoms are updated by blocks. My proposed method could benefit applications such as image denoising and audio track separation. Although this research area is new to me, I have shown the ability to quickly adapt to new knowledge and bring up novel ideas. Furthermore, I have procured precious experience in working collaboratively with my mentor in both experimental design and academic writing. 

After joining Huawei in Dec 2019, my research has focused on two parallel fields of study: network optimization and robotics, both involving solid mathematical theory and strong engineering practice. Working closely with my colleagues, I have solved problems in very large-scale optical network optimization. An exemplary work of mine was developing a hybrid algorithm that combined heuristics algorithm and mixed-integer programming to achieve world-leading performance in ODN network planning compared to rivals' products. I was awarded the 'Quality Star of Huawei' (Top 5%) for my work's excellency in both scientific innovations and engineering robustness. This experience gained me a deep understanding of mathematical tools in optimization and graph theory that are inherently the same as those in robot motion and task planning. Apart from network optimization, I have accomplished projects spanned across service robots that can help people with household tasks, to mobile manipulators that are designed for IT room maintenance. In addition, I have accomplished projects on visual servoing. I proposed a novel image-based pose stabilization control framework for holonomic mobile base with a fixed overhead camera. While doing these projects, I found myself with a stronger understanding of various topics of robotics study, including computer vision, SLAM, and optimal control. I was also for the first time, thinking about one ultimate problem: How to manipulate robots better, achieving dexterous locomotion and manipulation that is evolvable during interactions with the outside world. The craving for knowledge about unknowns has served as my strong motivator to pursue a PhD program. Besides R&D work, my experience as a CCSA (Chinese standardization association) representative has greatly honed my communication and interpersonal skills. I have led the industrial optical bus standardization project, working closely with the Ministry of Industry and Information Technology. My work requires intensive discussion with stakeholders from different backgrounds and writing white papers. This experience has given me valuable practice in communicating my ideas to colleagues and other members of the association in an attempt to procure support for my work. Moreover, I was able to think about scientific research holistically, evaluating impacts from both social and commercial points of view. Working in Huawei contributes several additional skills to my research skill-set, including proficiency in computer programming for engineering applications, time management skills and communication skills.

In summary, my past research experience reveals my interest in robotics and grant me the adequate skills working as a PhD candidate. Looking forward, I am excited to have the opportunity to conduct advanced-level research in Hong Kong.

**Reasons for wishing to pursue PhD studies in Hong Kong (in max. 2000 characters)**

Hong Kong is a fun and vibrant city where people from around the world come to live and study, attracted by its culture, heritage and its unique fusion of East meets West. It has some of the most well-known universities in the world and in particular, Asia. Studying in Hong Kong could enhance my understanding of my major from a global perspective while keeping the similarities in culture and closeness in geography to my home. Furthermore, the generous HKPFS scholarship would make my life comfortable and help me focus on my studies thus not to be worried about financial conditions. Benefit from Hong Kong's unique role in China's ambitious Greater Bay Area Plan, AI+Robotics has a promising future in Hong Kong's new trends of smart re-industrialization. I deeply believe that Hong Kong will maintain its economic competitiveness and social stability in the long-term future. And My specific field of research will have ample opportunities in Hong Kong. Apart from Hong Kong's privilege in educational and social aspects. I also love the environmental diversity that it offers. It is reputed for skyscrapers and bustling streets, but it has much more to offer. Hong Kong has several charming coasts located a stone's throw from the hustle and bustle of the city center, and it also owns several mountains with glamorous hikes. I am particular in favour of cities with spectacular views of both mountains and seas. All the factors mentioned above make Hong Kong the unique place for me to choose for my PhD studies.

 **Long-term career plan, aims and interests for future development after graduation (in max. 2000 characters)**

During the past years, I have seen many positive and negative circumstances in my life and profession. The most important lesson I have learnt is 'Do not live in other's dreams. There are two things that I am determined to in my entire life. One is to keep learning something new and interesting every day, another is to benefit the society. All my long-term plan, aims and interests will be circled around the two themes. I believe the doctoral program can open up new lines of inquiry for pursuing my career in robotics, and want to use it as the cornerstone for a successful career. My future plan is to launch my own start-up in the robotics industry, where I can express my passion for both technology and business as well as open my mind in ways that I could never do in any other form of occupation. As I am enthusiastic about transforming research outcomes into real-world products that bring positive impacts to our general society. To fulfil my aspiration, I aim to best harness the resources provided by my target universities and the city of Hong Kong, including learning from my predecessors, working assiduously for innovations, and meeting with venture capitals. Apart from professional occupancies, I am also interested in popularizing scientific and technological knowledge to the broader public audience, and serving in charities to help the less fortunate. Obtaining a degree does not secure me a promising future and there is much more that I should do to boost my future career. I will use all my determination, entrepreneurial spirit and persistence to create, innovate and succeed in my future.

**Contribution that the applicant would like to make to the development of research in Hong Kong and to society**

The first measure I would think of is to orientate my research goal to match areas of societal importance in Hong Kong. With its proximity to the Mainland, Hong Kong serves as a gateway for domestic and foreign AI+robotics companies to access the massive manufacturing base in the Greater Bay Area and other regions on the Mainland. Thanks to my past experience studying abroad in the UK and working as a professional in Shenzhen, I am able to understand people from diverse social and cultural backgrounds. Furthermore, I have a solid academic background as well as a business awareness of seeking the commercial benefits of scientific research. Hence, I foresee myself as the bridge between the Chinese Mainland and Hong Kong, and between academia and industry. I plan to play an acting role in collaborations with universities' industrial partners. My potential contribution would be transferring research outcomes to practical applications that meet the enterprise goals in both Hong Kong and mainland China. Certainly, working with elite professors during my PhD study will allow me to produce world-class papers to be published in the world's top journals and conferences, which add on to Hong Kong's academic reputation. Apart from making contributions to the research and industrial community, I would like to communicate science to the broader audience, whether it is my own research or general knowledge beyond my profession. This includes writing articles on social media (Zhihu, Twitter, etc.), participating in volunteer services and providing advice to elected officials.

**Other remarks in relation to the application for Hong Kong PhD Fellowship (in max. 2000 characters)**

For me, the situation is very different. I have been working since 1992 and have seen many positive and negative circumstances in the profession and for my career.

Each of you has a chance to discover who you are, rather than who you should be. A chance to live your dreams, not the dreams of others.



**Question 2. Proposed Research Topic - Please describe your proposed research topic (in max. 500 characters).**

Humanoid robots have vast applications (e.g. construction, households). Due to high-dimensional continuous state-action space and long-execution time, bimanual manipulation tasks are noticeably difficult. Imagine a humanoid robot has learned several discrete skills. Can it compose its skill library to a complete high-level task, with generalization to distinct environment settings? We intend to devise a bimanual manipulation learning scheme that can rapidly adapt to complex and long-term tasks.

**Question 3. Proposed Research Plan - Please describe your proposed research plan with detailed objectives and methodology for doctoral research (if any), please also provide references or citations, if applicable. (in max. 7500 characters).**
*The description should include the general field of the research; the specific research questions(s) or project aims; the research methodology or approach; the applicant's experience, participation in research projects, or preliminary results that support the feasibility of the work (if applicable); the significance, originality and/or anticipated impact of the work*

(1) Introduction

Imagine a humanoid robot has learned two discrete skills: 1) lifting bricks and 2) carrying bricks. Can it transfer its skill to a complete high-level task of carrying heavy loads from one point to another, or even recover from failures? And can it generalize its ability to different shapes of bricks? We intend to summarize the two inquiries into one fundamental question: How to design a bimanual manipulation learning scheme that can rapidly adapt to distinct state-space models, evolving during interactions with the outside world? Solving this problem will give us the key to real embodied intelligence and have profound benefits for robotics research.

(2) Literature Review

Bimanual Manipulation is a growing research topic in robotics. Recently, numerous research focus on long-horizon task-level skill learning. This not only needs to learn robotic bimanual manipulation skills in a continuous state-action space but also must learn a policy that decides which skills to use. Learn From Demonstrations (LfDs) methods such as Dynamic Movement Primitives (DMPs) (Ijspeert et al., 2013), Probabilistic Movement Primitives (ProMPs) (Paraschos et al., 2013) and Task-Parameterized Gaussian Mixture Model (Calinon et al., 2012) aim to learn an optimal robot control policy which is compliant to the distribution of trajectory demonstrations. There are various works on applying movement primitives to bimanual manipulations such as stir-fry (Liu et al., 2022), folding assembly (Almeida et al., 2016) and impedance dual-arm control (Batinica et al., 2017). However, all these approaches need a complete task description or strong human supervision. These assumptions limit the scalability of these methods in realistic tasks. Furthermore, the movements of the two robotic arms are often learned independently with two separate MPs, making it hard to handle the coordination between arms. Motor primitives could be combined with reinforcement learning for parameter tuning, using several rollouts to find a better policy update. An exemplary work is Policy Learning by Weighting Exploration with the Returns (PoWER) that combined reinforcement learning with motor primitives for parameter tuning based on expectation maximization (Kober and Peters, 2008). EM-inspired algorithm achieved good results on continuous tasks such as pancake-flipping (Kormushev et al., 2010), with only a small number of trials. However, this approach does not solve the problem of long-horizon bimanual manipulations, which require the robot to react to surroundings with a prolonged sequence of motor actions, which has a huge policy search space. In an effort to scale down the search space, Hierarchical Reinforcement Learning (HRL) treats learning the bimanual skills in a hierarchical manner, composing learned primitive movements via reinforcement learning (Xie et al., 2020). In (Nasiriany et al., 2022) and (Chitnis et al., 2020), the whole framework is decomposed into a state-independent task schema (sequence of skills) and a state-dependent policy. While these works achieve certain level of generality in long-time horizon, they did not consider much about the task-specific constraints. Therefore they cannot perform well on tasks that need close dual-arm coordination (e.g. one hand is holding an object while another is manipulating it). In addition, this set of methods usually need careful reward definition and subtask segmentation, making them hard to generalize.

(3) Proposed Methodology

We extend the idea of HRL further in this proposal, giving an overview of a possible approach to solving long-horizon, sparse reward tasks given a set of parameterized skills by learning a policy that chooses which skill to execute and what arguments to use when invoking it. In this proposal, we specifically focus on utilizing both learning-based methods (reinforcement learning) and model-based methods (movement primitive). We will first train several predefined movement primitives by imitation learning. The training trajectories can be acquired through Virtual Reality motion capture devices or motion retargeting from videos. Because each motion primitive has only limited scope in terms of duration and the tasks it can achieve. Therefore, after collecting a library of skills, we train two policy networks for both arms via reinforcement learning: A low-level policy networks determine skill parameters that automatically adapt to current continuous state-space arguments and a prespecified sub-goal. In addition, a high-level policy network is trained to decompose a long-horizon task into short sub-goals, which serve as input to the low-level policy network. Hence, the policy search space is significantly reduced, and we expect a better rate of convergence. To achieve tightly coupled bimanual coordination, we can use a Graph and Transformer Attention module to capture the Spatio-temporal relationship between dual-arm movements (Zhang et al., 2021) and between arms and surrounding objects. Furthermore, an intrinsic regularization term could be added to the RL reward function to avoid conflict between arms. By utilising this compositional structure of planning to generate diverse behaviours, we are able to reduce the difficulty of training a diverse goal proposal model.

(4) Limitations 

The state space parameter of the policy network is composed of high-dimensional observations and sub-goals. In domains with high-dimensional observations such as images, this may require explicitly optimizing over image pixels, which is undesirable. Hence it is crucial to find latent expressions of the image signals, similar in (Kingma and Welling, 2014). For generality, the lower-level controllers should be supervised with goals that are learned and proposed automatically by the higher-level controllers (Nachum et al., 2018), without any human prior knowledge. Hence we shall look into the mechanism of segmenting the overall goal into meaningful subtasks.

(5) Expected Results

We plan to evaluate the proposed method on variations of bimanual tasks both in simulation environment and on physical robots. Use opening bottle as an example, the goal is to rotate a cap while keeping the bottle static in hand. The generalization of our method could be tested with different bottle starting locations, and varies the cap's coefficient of friction. We can evaluate the results in multiple aspects, including the rate of convergence and the success rate of execution in both seen and unseen scenarios.

(6) Conclusion 

Bimanual robotics systems have shown a promising future in many manufacturing, assembly lines, and household processes. Our work demonstrates the potential to take this intelligence one step further, that utilising fully automated policies for complex and long-horizon tasks by presenting a multi-level reinforcement learning framework and neural network encoding of bimanual constraints. We expect to obtain results that excel the baseline and we intend to investigate the idea further in future studies.

*Detailed version RP and full bibliography can be found in uploaded documents*

## /My Application Documents/Personal_Statement_NUS_.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Personal_Statement_NUS_.pdf

## /My Application Documents/Personal_Statement_UIUC_.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Personal_Statement_UIUC_.pdf

## /My Application Documents/Research_Proposal_on_HRL.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Research_Proposal_on_HRL.pdf

## /My Application Documents/Research_Proposal_on_TAMP.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Research_Proposal_on_TAMP.pdf

## /My Application Documents/Source Code/Academic Statement of Purpose(UIUC).zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/Academic Statement of Purpose(UIUC).zip

## /My Application Documents/Source Code/Personal Statement(NUS).zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/Personal Statement(NUS).zip

## /My Application Documents/Source Code/Personal Statement(UIUC).zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/Personal Statement(UIUC).zip

## /My Application Documents/Source Code/Research Proposal on HRL.zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/Research Proposal on HRL.zip

## /My Application Documents/Source Code/Research Proposal on TAMP.zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/Research Proposal on TAMP.zip

## /My Application Documents/Source Code/recommendation-letter-chen.doc

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/recommendation-letter-chen.doc

## /My Application Documents/Source Code/recommendation-letter-lu.docx

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/Source Code/recommendation-letter-lu.docx

## /My Application Documents/recommendation-letter-chen.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/recommendation-letter-chen.pdf

## /My Application Documents/recommendation-letter-lu.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/My Application Documents/recommendation-letter-lu.pdf

## /Personal Statement/ShortPersonalStatement.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/ShortPersonalStatement.pdf

## /Personal Statement/StatementOfPurpose-MIT.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/StatementOfPurpose-MIT.pdf

## /Personal Statement/StatementOfPurpose-UCB.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/StatementOfPurpose-UCB.pdf

## /Personal Statement/UMPersonalStatement.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/UMPersonalStatement.pdf

## /Personal Statement/personal_statement.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/personal_statement.pdf

## /Personal Statement/ps-cal-eylai.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Personal Statement/ps-cal-eylai.pdf

## /Practical Experience/浙大飞跃手册.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Practical Experience/浙大飞跃手册.pdf

## /README.md


# PhD-Application

### CV, Resume
- [How to write a EECS Resume](https://www.eecsresume.com/article)
- [Template on Overleaf](https://www.overleaf.com/latex/templates/resume-template-by-anubhav/dhmkrwtksdgy)
- [moderncv](https://ctan.org/tex-archive/macros/latex/contrib/moderncv?lang=en) [github repo](https://github.com/xdanaux/moderncv)

### Personal Satement
> Candidate should describe succinctly your reasons for applying to the programme, your preparation for the field of study, future career plans, and other aspects of your background and interests that may help in the evaluation of your aptitude and motivation for graduate studies.

- [A Guide from MIT](https://mitcommlab.mit.edu/broad/commkit/graduate-school-personal-statement/)
- [A Sample from Christopher's Blog](http://cwfletcher.net/Pages/SoP.php)
- [A Five-Minute Guide to Ph.D. Program Applications](https://pg.ucsd.edu/PhD-application-tips.htm)
- [How to Write a Bad Statement for a Computer Science Ph.D. Admissions Application, CMU](http://www.cs.cmu.edu/~pavlo/blog/2015/10/how-to-write-a-bad-statement-for-a-computer-science-phd-admissions-application.html)
- [An Example from MIT CSAIL](https://mitcommlab.mit.edu/eecs/wp-content/uploads/sites/6/2016/09/CS-grad-school-personal-statement-annotated-example.pdf)
- [An Example from Eugenie Y.Lai](https://eugenielai.github.io/docs/application/ps-cal-eylai.pdf)
- [Jiayi Weng's PS](https://trinkle23897.github.io/pdf/PS-CMU.pdf)

### Academic Statement of Purpose (SOP)
> This statement should include your goals and background as it relates to the field of computer science. It is good to include any past work history, your interests, and reasons for pursuing a graduate degree. Your statement will be evaluated for your conciseness, reasons for pursuing a graduate degree, leadership skills, and eloquence.

- [Writing the Statement of Purpose, UCB](https://grad.berkeley.edu/admissions/steps-to-apply/requirements/statement-purpose/)
- [How to Write a Statement of Purpose for Grad School, Swapneel Mehta (NYU)](https://swapneelm.github.io/how-to-write-a-statement-of-purpose-for-grad-school)
- [Example from  Anant Bhardwaj, MIT](http://people.csail.mit.edu/anantb/public/docs/sop/phd_admission_sop.pdf)
- [Sample from Eugenie Y. Lai, MIT](https://eugenielai.github.io/docs/application/sop-mit-eylai.pdf)
- [A very useful tutorial for me, Eugenie's Blog](https://eugenielai.github.io/posts/another-annotated-sop.html)
- [Writing a Statement of Purpose, Swapneel’s Blog](https://djunicode.github.io/2018/10/16/writing-a-statement-of-purpose.html)
- [Official Example from MIT](https://mitcommlab.mit.edu/broad/wp-content/uploads/sites/5/2016/09/Broad_CommKit_GradSchool_Personal_AAE2.png)

### Miscellaneous between SOP and PS
> To me, the major difference between PS and SOP is that Personal Statement (PS) requires a frameworked story about your life which illustrates your future career goal, your social awareness and your potential to contribute to a community. A Academic Statement of Purpose (SOP), on the other hand focus more on your research interest and past research experience. Read each program’s prompts and formatting requirements carefully. The point above is about the content, while this one is about the format and separation of the content. Some programs don't differentiate SOP and PS, but require you to submit one document that illustrate both your personal side and academic side. However, programs like Berkeley EECS require an SoP and a PS so the separation depends on the prompts.

### Research Satement and Research Proposal
> Research Statements from professors/research scientists/Ph.D. Students often yield insight into not only their goals, but also areas of active research in general which is why I’m including some samples for candidates to read in order to ‘get a better idea of their field’ before writing their Statement of Purpose.
- [Samples from UNC Champel Hill](https://www.cs.unc.edu/~cssa/guides/proposals/index.html)
- [A Guide from UCL](https://www.ucl.ac.uk/ioe-writing-centre/plan-your-assignment/write-a-research-proposal)
- [A Guide from Imperial College London](https://www.imperial.ac.uk/research-and-innovation/research-office/preparing-and-costing-a-proposal/planning-a-proposal/)
- [A Guide from CUHK](https://www.ilc.cuhk.edu.hk/EN/ENResources/Writing_Research.aspx)
- [A Tutorial from Zhihu](https://zhuanlan.zhihu.com/p/298343066)

### Recommendation Letter
- [A useful Latex Template](https://github.com/Dieselmarble/PhD-Application/blob/main/Recommendation%20Letter/THU%20Letter%20of%20Recommendation%20Template.zip)

### Contacting Potential Supervisors 
- [Advice from UCL](https://www.ucl.ac.uk/prospective-students/graduate/sites/prospective-students_graduate/files/potential-supervisor.pdf)

### Search Tool
- [CS Rankings](https://csrankings.org/)

### GRE
- [A list of Universities that (not)require GRE](https://github.com/Dieselmarble/PhD-Application/blob/main/GRE/CS_CS-Adjacent%20GRE%20Requirements.xlsx)

### Community BBS and Blogs
- [1Point3Acres 一亩三分地](https://www.1point3acres.com/bbs/)
- [gter 寄托论坛](https://www.gter.net/)
- [A Guide from Hongzheng Chen](https://chhzh123.github.io/blogs/2021-05-22-phd-application/)
- [Collection of Experience from ZJU](https://github.com/Dieselmarble/PhD-Application/blob/main/Practical%20Experience/%E6%B5%99%E5%A4%A7%E9%A3%9E%E8%B7%83%E6%89%8B%E5%86%8C.pdf)
- [Jiayi Weng's Blog](https://trinkle23897.github.io/posts/application#%E6%9C%AC%E7%A7%91%E5%AD%A6%E6%A0%A1%E9%99%A2%E7%B3%BBgpa)
- ["Strong Committee" and "Strong Professor" in admission, Zhihu](https://zhuanlan.zhihu.com/p/128037373)


## /Recommendation Letter/THU Letter of Recommendation Template.zip

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Recommendation Letter/THU Letter of Recommendation Template.zip

## /Research Statement/A Topological Framework for Visualizing Time-varying Volumetric Datasets.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Research Statement/A Topological Framework for Visualizing Time-varying Volumetric Datasets.pdf

## /Research Statement/Research_Statement_DM.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Research Statement/Research_Statement_DM.pdf

## /Research Statement/research_statement_UIUC_CS.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Research Statement/research_statement_UIUC_CS.pdf

## /Statement of Purpose/UMStatementofPurpose.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Statement of Purpose/UMStatementofPurpose.pdf

## /Statement of Purpose/phd_admission_sop.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Statement of Purpose/phd_admission_sop.pdf

## /Statement of Purpose/sop-mit-eylai.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Statement of Purpose/sop-mit-eylai.pdf

## /Statement of Purpose/statement-of-purpose.pdf

Binary file available at https://raw.githubusercontent.com/Dieselmarble/PhD-Application/refs/heads/main/Statement of Purpose/statement-of-purpose.pdf


The better and more specific the context, the better the LLM can follow instructions. If the context seems verbose, the user can refine the filter using uithub. Thank you for using https://uithub.com - Perfect LLM context for any GitHub repo.