Go to the content anchor
Visitor:174269
:::
Visitor:174269
:::
:::

2020 MOST GASE Taiwan GTI Program Experience Sharing

MOST GASE Taiwan GTI Program Experience Sharing(2020)

By Jhalak Gupta

Indian Institute of Technology,Indore, India

 

After two months of this virtual internship program, I appreciate the organizer, MOST GASE, my supervisor, Prof. Jane Yung-Jen Hsu and the team for their efforts to give me the chance to learn even during this challenging pandemic period. My experience of the internship at GTI Program has been prodigious. Starting from orientation week, which included a wonderful Opening Ceremony in which I got to meet everyone and had fun icebreaking activities. Then there were insightful talks on Artificial Intelligence by Mr. You-heng Richard Pei and Green Tech by Ms. Y.H. Horng through which I got to learn about the prospective development of AIoT and energy transition. Apart from these, there was an enjoyable and interactive course on Mandarin and lastly an excellent virtual tour of Longshan Temple to give us the experience of Taiwan culture even from our home. 

 

Those six weeks which I got to work on the exciting project on "Artificial Intelligence for infant motor screening" were full of learning and exploration. In this research project, our objective was to establish a solution for infant movement analysis using AI approach and techniques. The physicists have several milestones and checklist items that an infant should perform. This assessment allows physicists to see the interplay of the child's neuromotor system within the specific physical contexts of the motor task and helps to identify if the infant has a delay in their growth. Through this project, we aimed to suggest the best yet the suitable architecture of AI model to investigate and analyze the infant's development based on their captured movements which will help the physicist in the physical therapy department to determine whether or not an infant performs the proper motoric movements.

 

My supervisor, Prof. Jane Yung-Jen Hsu and the mentoring team were amazing and made me feel so welcome. They were incredibly supportive and helpful. I had my final exams of college during the initial two weeks and was a bit worried about how I will manage, but they were so understanding and didn't burden me at all. We had weekly meetings to sync up in which everyone working on different projects gave updates and had discussions to provide me with an experience of the process of research and how everyone works. Also, they always gave me a chance to speak what I feel and welcomed my views. They even entertained my silliest doubt. I had a few problems and challenges as well due to working remotely like lack of computational resources and the difference of time zones. But everyone tried to help as much as possible. With the help of MOST GASE team, I also got the opportunity to visit the Intelligence Agent Laboratory virtually and see the demonstration of the working of the project. 

 

Seeing the end of this incredible internship was hard but was done with such a beautiful closing ceremony in which everyone shared their work and experience. MOST GASE team also made a video which included everyone through small clips sent by us and was so joyful to watch all of us brought closer when miles apart. Throughout the internship, there wasn't a single moment in which I felt like an outsider. Everyone who I met during this time was so helpful and wonderful. As this project was in its preliminary stage, the collection of the actual dataset consisting of video clips of different milestones of infants was in process. So I wasn't able to evaluate the constructed architecture on the actual dataset. But I have been fortunate to get a lot of learning despite COVID-19, and I'd like to explore opportunities to continue further my research under the supervision of Prof. Jane Hsu. Even though I didn't get a chance this time, I hope one day I can visit this beautiful little island – Taiwan in the future, to see my supervisor, lab and my friends in GASE.

Last Modified : 2020/10/19