Feel free to answer as many questions as you like, but it would be great if everyone answered everything! These survey answers will be included in the 2015 competition report, so the more info the better. Please feel free to provide external references/links as necessary Bot Name: Tyr Bot Race: Terran Author Name(s): Simon Prins Affiliation(s): - Nationality(s): Netherlands Occupation(s): Software Developer (These will be listed on the competition website) Bot URL: https://github.com/SimonPrins/Tyr/wiki Personal URL: - Affiliation URL: - Q: How did you become interested in Starcraft AI? I saw a post about the sscaitournament on the Day9 forums. I really liked the idea of programming a Starcraft AI, so I started creating my own bot Q: How long have you been working on your bot? I started just after the 2014 sscaitournament, so about ten months. Q: About how many lines of code is your bot? Around 9000 Q: Why did you choose the race of your bot? I like playing Terran in Starcraft II. However, I like all races, so the bot is designed so that it can also play other races. I hope to enable my bot to play other races in the future. Q: Did you use any existing code as the basis for your bot? If so, why, and what did you change? I did not use any existing bot. I did the tutorial for creating a java bot and started from there. Q: What is the overall strategy/strategies of your bot? Why did you choose them? Mainly it uses a lot of mech and goes for a longer game. I started out by training a lot of marines, but I found that they died very quicly, so I switched to a more sturdy mech army. Q: Do you incorporate learning of any form in your bot? If so, how was it accomplished? My bot does not use any learning. It does select specific strategies against specific components, but this is based on hard coded decisions I made myself. Q: Do you use any interesting AI techniques or algorithms in your bot? If so, which? I use an agent based architecture, both for controlling individual units and for controlling groups of units. Other than that, I do not use any specific techniques. Q: What do you feel are the strongest and weakest parts of your bot's overall performance? My bot has a solid defense, which is difficult for weaker bots to penetrate. It can use different strategies against different bots, however these are hardcoded, which is a big downside. Some weak points I am hoping to fix in later versions: -My bot does not keep its army together, so units trickle into a fight one by one. -My bot does not defend far away expansions. -There are many small flaws where my bot makes bad decisions. These do not occur very often and they will not cost it the game by itself, but by I smart opponent they can be exploited. Q: If you competed in previous tournaments, what did you change for this year's entry? The only previous tournament my bot competed in was the IEEE CIG 2015. I have not had much time to update my bot since then. Mostly I changed the strategy selection so that my bot will use better strategies. All these strategies already existed in that version of the bot, though. Q: Have you tested your bot against humans? If so, how did it go? I have played against my bot once or twice, but I found it very easy to beat. I have not played the latest version though, so I expect that it will be slightly harder to beat now, but still not too difficult. Q: Any fun or interesting stories about the development / testing of your bot? - Q: Any other projects you're working on that you'd like to advertise? - Optional Opinion Questions: Q: What is your opinion on the current state of StarCraft AI? How long do you think before computers can beat humans in a best-of-7 match? Compared to humans, the AIs are still very weak. All bots have weak points, where they make bad decisions, which can be exploited by players. I think it will be around 5 years before bots can beat humans in a best of 7 match. Q: What do you feel is the biggest hurdle (technological or otherwise) in improving your bot's AI? I feel that map awareness is very difficult to achieve. With better map awareness, things like good army positioning could be much easier to achieve. Q: Which bots are the most interesting to you and why? Tscmoo and Tscmooz are both very powerful and interesting to watch. The same goes for some of the new zerg bots, for instance Killerbot, ZZZBot and Overkill. ICEBot is very interesting since it is very well-rounded, however a lot of bots have figured out how to beat it using early rushes.