Ryan S
- Research Program Mentor
MEng at Massachusetts Institute of Technology
Expertise
Deep Reinforcement Learning, Computer Vision, Machine Learning, Neural Networks, Perception
Bio
I recently finished my Master's at MIT, where I studied artificial intelligence and conducted research in MIT CSAIL's Distributed Robotics Laboratory under Professor Daniela Rus and Professor Sertac Karaman. I completed my undergrad in electrical engineering and computer science and mathematical economics at MIT in 2020. My main Master's research was at the intersection of deep reinforcement learning, computer vision, and robotics. Specifically, I worked on improving the sample efficiency of deep reinforcement learning algorithms for autonomous driving applications. I am also interested in the fields of optimization, game theory, and electromagnetics. Outside of MIT, I now work as an image scientist, and I enjoy Krav Maga, weightlifting, biking, and dark roast coffee. I have recently discovered my passion for teaching and mentorship, as well as technical writing. Finally, I would not be who I am without my favorite playlists on Spotify!Project ideas
Training a Deep Reinforcement Learning Agent with TensorFlow Agents and OpenAI Gym!
Deep Reinforcement Learning is a rapidly growing field at the intersection of machine learning and robotics, and through this project, you will have the chance to implement deep reinforcement learning agents yourself! We will implement this project using TensorFlow-Agents, a powerful library for training agents to solve tasks such as getting a Half-Cheetah to walk, balancing an Inverted Pendulum, or teaching a car to race autonomously (all in simulation).
Fingerprint Classification and Pattern Recognition in Keras
Ever wondered how fingerprint scanners work? In this project, we will study and implement some ways in which fingerprints can be detected and identified using deep learning (implemented through Keras in Python) and classical computer vision techniques Iimplemented through OpenCV in Python).
Computer Vision for Autonomous Driving!
Fast and accurate vision is a crucial component for the successful adoption of autonomous vehicles as a widespread technology. In this project, we will use Python and PyTorch, a deep learning library, to segment different objects using RGB cameras and lidar sensors!