Gym Car Racing Environment
- Used transfer learning to achieve the fast lap times in OpenAI’s Car racing environment by training the agent on one circuit, and racing it on other customized target environments by zero-shot transfer or by additional fine-tuning.
- Compared the performance of model-based and model-free approaches, and observed that model-based approaches dominate in performance and converge faster than model-free approaches in this environment.
- Observed that transfer learning in most setups not only boosts the performance on the target domain, but also shows high performance ability during learning.
Benedict Florance Arockiaraj
ML Engineer
My research interests are at the juncture of deep learning and computer vision.