The goal of the project is to train an agent that plays the ‘Dangerous Dave’ game. The objective is to prevent any human intervention and make the agent learn the optimal moves. This can be achieved by utilizing various reinforcement learning and deep learning techniques. There are various levels/stages in this game but the scope of this project is to utilize the 1st stage of the game for training and testing purposes.
Akash Patel (akashkum@usc.edu) Harima Patel (harimapa@usc.edu) Jainam Divyesh Shah ( jainamdi@usc.edu) Mansi P Shah (mansipra@usc.edu) Prabhat Mishra (prabhatm@usc.edu) Yash Chanchad (chanchad@usc.edu)