highlights SMAWM Structured Multi-Agent World Models is my research project done under CS4246 AI Planning and Decision Making. BTRL Barlow Twins for Model-based Reinforcement Learning is my 1-year undergraduate research project advised by Prof. Harold Soh at the Collaborative, Learning and Adaptive Robots (CLeAR) AI Lab. Slime-RL Bayesian Multi-Agent Reinforcement Learning for Slime Volleyball is my CS3244 Machine Learning class project that won 1st place at the NUS Computing Project Showcase, 17th STePS 2020. class The Chosen One Multi-agent reinforcement learning trained by adversarial training, was awarded the Best AI Hack at the iNTUition Hackathon 2019 at Nanyang Technological University. COVID Policy Simulator An agent-based modelling (ABM) study on mask wearing and its effect on policy interventions, done for my USP quantitative reasoning module. Idle Trading Hero An automated algorithmic trading platform that uses quantitative analysis, was my Orbital project in summer 2020, awarded Artemis, the highest level. work hires needed Feature experimentation for Indeed's match recommendation platform is my internship project at Indeed in summer 2022, which improved matches for 300M job seekers. Simkit A framework for building generative and probabilistic models for training reinforcement learning agents, is my internship project at Grab in summer 2020, which I presented at Google Developer Space. DrFAQ An open-source question answering chatbot, that combines a series of frequently-asked question (FAQ) similarity matching, deep natural language processing (NLP) question answering and search. Chiller Doctor A time series anomaly detection algorithm, designed and implemented during my internship at Infocomm Media Development Authority in 2019. fun CARElytics An AI-enabled social media analytics solution for real-time brand and product insights, was designed during the HACK Asia Global Hackathon 2020, and was awarded Top 10 Finalist. Optigram A data-driven interactive article on climate change, awarded 1st Runner Up for the Optigram Data Visualisation Competition 2021. This Gemstone Does Not Exist A deep convolutional generative adversarial network (DCGAN) for generating gemstones, was done as part of my CS6101 guest lecture class project, presented at NUS Computing Project Showcase in 2019.
