School of Engineering Department of Computer Science and Engineering 87 VR Metaverse for Education Supervisor: BRAUD, Tristan Camille / CSE Student: YE, Weicheng / SENG Course: UROP1000, Summer This is a progress report for individual projects, which mainly focus on creating a Chatgpt assistant which can interactively chat in virtual reality. To be specified, it can be divided into 3 components—-Specific, Measurable and Time-bound. For “Specific” part, this project, which may last for 2 or 3 months, will focus on creating Chatgpt assistant with character models in Unity, modifying the actions of models and setting interactable AI in Virtual Reality room. To achieve the above, all I need to do is to implement the following functionality ——Text-to-Speech, Speech-to-Text, Chat Gpt installation, Movements of anime characters and VR interactions. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN, Gary Shueng Han / CSE Student: ABDUL REHMAN / SENG Course: UROP1100, Summer Extracting useful results from raw data has been the focus of much research over the last decade. One such form of raw data is Irregularly Sampled Time Series (ISTS) data. In this study, we elaborate on the use of ISTS data in forecasting and classification tasks. Additionally, we evaluate different deep learning-based approaches used to perform said tasks, and test their performance on publicly available BTC-USD exchange rate and PhysioNet2012 ICU Patient Survival datasets. The report is divided into three parts each dedicated to a specific problem. We discuss Time Series Analysis in Part I, ISTS forecasting in Part II, and ISTS classification in Part III. The models discussed in the report are available at UROP1100 Implementation. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN, Gary Shueng Han / CSE Student: CHI, Yankuan / SENG Course: UROP1000, Summer Machine learning gains increasing attention in recent years and has a great number of applications nowadays. It has become one of the most important parts of computer science. In this UROP course, I explored some basic concepts and models of machine learning regarding computer vision. I tried out three models, where the first one, a sample model from Pytorch, is for image classification and the last two, MCNN and CSRNet, are for crowd counting. For all three models above, I tried studying some aspects of them, as well as making some changes to them to see how their performances change.
RkJQdWJsaXNoZXIy NDk5Njg=