Scientifc computing/ CSE master PS求批评

UPENN
I recall that the trickiest part of the coursework assignments was how to show weather elements via the processing of data. Although I was always keen on becoming an atmospheric scientist, it finally dawned on me that I should broaden my knowledge framework, combining my undergraduate major with the most potent tool–computer science, when I was writing my paper on air quality variation during COVID-19 this summer. During this research, I need to visualize a deluge of data: four months of hourly air pollution data obtained from 367 prefecture-level Chinese cities and about from 2000 observatories. With limited coding experience and a tight schedule, I was propelled to learn Python by browsing various tech blogs and videos each day. Solving problems directly from scratch was difficult. Without the help of unknown meteorologist bloggers I could never generating out all the data and figures I need. However, I was shocked to hear that those bloggers all turned to work as AI Engineers in the IT industry and conducted challenging projects beyond the area of Atmospheric Science. The more I learned the more I realized that AI is irreversibly the future trend of the world. I must equip myself with solid computer and data science knowledge to join the next scientific revolution, which is also vital for advancing the development of Atmospheric Science. Then I began to action.
I was supposed to be in the apartment of University of Manchester this semester, where I was not allowed to pick courses in CS department. Nevertheless, I could stay in China since all activities in Manchester went online due to COVID-19 Since. I took the chance to took six courses in CS department at my original university in Nanjing in Fall, 2020. It was hard for a student from another discipline without adequate perquisite knowledge like C programming to catch up the lecturers’ progress, but I successfully made it by learning and practicing through Internet sources like Stackoverflow and w3school, especially proficient in C++. At the same time, I sought a Data Intern occupation in the Meteorological Research Institute, responsible for cleansing, gridding and visualizing meteorological elements. The official API interface always encountered time out when downloading data under the Institute’s network environment, I rewrote some parts of the codes, which solved the error and improved the team's work efficiency. Moreover, as a final year student, I began to look for a graduate program that can best integrate with Atmospheric Science and equip me with powerful skills to enter the job market as a competitive player.
The interdisciplinary training provided by the MSE Scientific Computing at the University of Pennsylvania is undoubtedly the perfect choice for me in obtaining the computational science ability for future progress in atmospheric science.  And I believe I was competent for finishing the degree successfully. Recalling the past four years, I have grown and matured. In my freshman year, with little interests in the fundamental courses, my grades were average. However, this situation was immediately improved at the start of junior year as I was fascinated by the core courses and realized the importance of foundational physics and mathematical courses, which won me the University First-class Scholarship eventually. My solid background in Computing, Mathematics, and Physics can be proved by grades in relevant courses and multiple projects during my undergraduate studies. I also possess a critical learning ability: in a hail storm modeling task, I modified the teacher’s manual and wrote a configure script for the class since the manual often ran into error. When studying at the University of Manchester, writing analysis reports based on the measured data made up the curriculum's central part, which improved my capability in data analysis and English reading and writing.  
Moreover, as the lead author of my summer research, I learnt to master Python to web crawl, to process various format of data and to visualize proper scientific graphs. The atmospheric background also made me sensitive to unnoticeable abnormal data features, redirecting me to update codes to generating graphs more fitting to the ground-truth. Besides, In the process of revising and adjusting research plan with my supervisor and other members, my critical thinking. academic writing and cooperation were significantly strengthened.
Currently I am doing my final year project, which was initially about synoptic meteorology. But I adjusted it to be more AI related: machine learning adaptations in precipitation prediction. since I made my decision to focus on the more interdisciplinary field. Thanks to my solid mathematic background, I successfully understand how the convolution and recurrence neural network work after some struggling reading, and planned to tried new structures like Attention and Transformer to improve the prediction accuracy.

Through the research experience in both machine learning and atmospheric science, I recognized that atmospheric science lacks roadmaps that exists in the Computer Science field, which causes beginners to take detours in the process of learning. Thus I launch a project to build knowledge map in the field of meteorology to provide learning guide for beginners in different areas in each subdivided field with several undergraduate students from both atmospheric science and CS major to of using Natural language.

All those exercise permits me with adequate knowledge to can perform good in your program.
Furthermore, the program's excellent core curriculum fits well my expectation to strengthen the hard skills in computer science. Courses in Programming Languages and Algorithm will significantly enhance my coding and algorithm ability, and courses in Big Data and Machine Learning provide me the key to the new era of AI and big data. My solid foundation for Fortran, Fluid Mechanics and Weather Forecasting modelling knowledge prepared me well for Bucket A’s Computational Mechanics and Finite Element Analysis after which I could have a stronger knowledge of fluid and heat transfer problem in numerical modelling. For the other 2 course units, I prefer an independent project, with which I plan to Applying scientific computing to natural science problems matches well with my background.  It is in line with my future career development goals. I plan to seek a job in IBM to use knowledge learnt from this program to empower the current atmospheric science tool: combining state-of-the-art techniques like artificial intelligence and numerical modeling to improve tornado prediction accuracy, which would save huge cost of false alarm and save lives from missing warn. I believe that everything, especially the systematic training of practical computing knowledge I obtain from the MSE Scientific Computing program, will undoubtedly lead me to the path I am passionate about.

可以，不错的嘛

西西MM 发表于 2020-11-11 09:35:32
可以，不错的嘛

貌似很不学术被中介批了😢

戴帽子 发表于 2020-11-11 18:54:55
貌似很不学术被中介批了

CS以工作导向吧，学术了干嘛