![]() ![]() There are some issues with behavioral cloning The inputs are images so we do not have access to some things like speed Data limitations as well so does not know what to do when car veers off track Policy based off human decisions, so cannot achieve superhuman performance or recognize even more complex patterns Good start Move on the reinforcement learning algorithm. ![]() Behavioral cloning mimics what a human would do Point out different portions of image (input, output, whole as data) Data: has the computer remember decisions at each time step Accuracy ![]() State Space: All possible states or configurations our self-driving car can be in Action space: Possible actions the self-driving car can be in -> Direction, step on the gas, break Rewards: if the car is doing what we want it to do, that is a reward
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |