However, it does not get access to the state of the puck or opponents during the gameplay. The agent also sees an image for each player, from which it can infer where the puck is and where the other teams players are. Image based agentsĪn image-based agent gets access to the state of the players karts. However, they have different inputs and performance requirements.įor each we play 2 vs 2 tournaments against agents the TAs and I coded up. You need to pick one route and cannot compete in both.īoth agent types have the same objective, score as many goals as possible and win the match. You have two options: An image-based agent and a state-based agent. Test your code on a Linux machine (either CS lab or colab).The task is to program a SuperTuxKart ice-hockey player.15 pts extra credit for the top 3 teams per agent type in the competition (15pts winner, 10pts second, 5pts third). Structure of the writeup, exposition, how well is the idea motivated and explained. We will compute the average against the TAs agents. 30pts average number of goals scored per game.How many things did you try? Do you understand how your strategy works? Please note that the grades posted by this grader will not not be your final grade on the project. Submit your codes under the ‘Final Project’ assignment and the sanity-check grader will automatically trigger like the rest of your assignments. We provide a grader on canvas for you to sanity-test your codes. The submission with the most victories wins the tournament.Īdditional, matches further break ties. Should both teams violate the rules, the game is tied.ĭepending on the number of submissions, we will let all submissions play against each other, or we will first have a group stage and then have the top 8 submissions play against each other. Should one of the teams violate the above rules, it will lose the game. Two of your agents will play two opponents.Įach game will play up to 3 goals, or a maximum of 1200 steps (2 minutes). We will play an ice hockey tournament between all submissions within each agent type. If you violate the time-limit your forfeit the match. You may use pystk and all its state during training, but not during gameplay in the tournament.Any attempt to circumvent this will result in your team failing. In our grader, your agent will run in complete isolation from the rest of the system.We will provide some test agents to compete against towards the end of the project.You may work in a team of up to 4 students.This project is completely open-ended, however, there are a few ground rules: The most important include recording a video -r my_video.mp4 or saving data -s my_data.pkl. The tournament runner has quite a few arguments you can explore using the -h flag. You can test your agents against each other or the build-in AI The current agents will simply drive straight as an example. This class provides an act function, which takes inputs as described above and produces a list of actions (one per player). In both bases you will implement Team object in player.py. Time limit 10ms per step (call to the Team.act function) on a reasonable fast GPU.Your agent needs to be a single deep network, no hand-designed components are allowed.We anticipate to see some learning-based solution here.Ĭhoose this path if you prefer to work on the control component of your agent. Time limit 50ms per step (call to the Team.act function) on a reasonable fast GPUĪ state-based agent gets access to the state of the players karts, the opponents’ kart, and the location of the ball.You may use any controller you want (hand-designed or even some of the agents the TAs and I coded up).You need to use a deep network to process the images.This option may require more compute to develop and test. We anticipate to see some vision-heavy solutions here.Ĭhoose this path if you prefer to work on the vision component of your agent and use a hand-tuned controller. The task is to program a SuperTuxKart ice-hockey player.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |