There are about 2 hours of lectures weekly. Most students reported spending up to 30 - 40 hours weekly, and even then, some still could not complete the projects. In terms of workload, this class is pretty heavy.
For example, their behavior on a website (where each click gives more information about what they will click next), their health (where each doctor’s visit or activity provides information on their overall health), etc. I was intrigued by model-free methods and thought it might be an interesting and practical approach to model aspects of people in the world, without a complete model of human characteristics. We learnt about model-based methods such as Value Iteration and Policy Iteration, as well as model-free methods such as Q-learning. =( Why take this course?Īt the tail end of the Machine Learning class I took last fall, a small introduction to reinforcement learning was made. As you can see, it crashed right onto the surface. Here’s how the agent did on its first try (in reinforcement learning, we refer to the “models” as agents more here).
This was implemented via deep reinforcement learning approaches. I highly recommend this course for anyone who’s part of the Georgia Tech OMSCS.Īn especially fun project involved landing a rocket in OpenAI’s LunarLander environment. Throughout the course, we learnt techniques that allow one to optimize outcomes while navigating the world, and were introduced to several seminal and cutting edge (at least back in 2018) papers.
#OMSCS DID I CLICK ACCEPT FREE#
Any free time I had outside of that was poured into the Georgia Tech Reinforcement Learning (CS7642), which is the subject of this post. Things have been super hectic with project Voyager at Lazada, switching over to the new platform in end March and then preparing for our birthday campaign in end Apr. I know, I know, I’m guilty of not writing over the last four months. Or view all OMSCS related writing here: omscs. You might also be interested in this OMSCS FAQ I wrote after graduation.
0 kommentar(er)
