6

Firstly, I'd try a simple greedy AI strategy that just goes to the nearest unexplored planet and explore it. It could be enough to make the game interesting while not making the AI invincible. With that said, supposing you want to practice ML, or want it just for fun, there are a few options: Store the input-output vectors you mentioned from human game-...


3

As @davidvanbrink said in a comment, gating is a common way of doing this: either put the starting position in an large enclosed area that requires all the skills to escape and contains basic tutorials on each, or have certain portions of the game be blocked by barriers that require one of the skills and make sure that every critical path contains one of ...


2

In general you want to give the neural net a reward function that it is trying to maximize, or, alternatively, a cost function it is trying to minimize. In the car example, "distance along the track" is a good indicator of progress. You can simply reward the neural net for making progress along the track. In the simple case, imagine a neural net that tries ...


1

every application works with data. both unity and tensorflow work with data. tensorflow is just a machine learning library that basically is made for python language. you dont need to find a wrapper to let you work with tensorflow but i think there is some. assume you need neural network as supervised learning system to teach you agent to move to right ...


1

Here is a good start http://unity3d.com/learn http://forum.unity3d.com Virtual Game Lab also have an updated list of resources and tutorials for learning Unity. http://www.virtualgamelab.com/unity-resources.html My recommendation for you is to pick a couple of old school 2d games and try to remake them. This will make you face real game development ...


1

In Artificial Intelligence for Games 2E the authors discuss racing games regularly and say that approaches built on steering behavior primitives (i.e. reactive planning) can be very well-suited to the domain. (I'm in the middle of reading it now, so I can't say whether they make the same claims for reinforcement learning.) It of course depends on your exact ...


1

Achievements. After learning one skill from set, display progress information like "Skill learned: 1 / 4" (eg. "Lighting fire: 1/4 skills learned"). This way player will know there are 3 other skills to learn.


1

To answer the three of your questions with one answer: sure you could, but you wouldn't be able to use the default content manager. However, you can create custom content processors to output file data in a format that only you would immediately know. This would still allow you to use the default XNA content manager (this.Content in any Game-derived class) ...


Only top voted, non community-wiki answers of a minimum length are eligible