AIIDE StarCraft AI Competition - Survey Feel free to answer as many questions as you like, but it would be great if everyone answered everything! Many people are interested in learning as much as possible about the bots that competed! Please feel free to provide external references/links as necessary Bot Name: XiaoYi Bot Race: Terran Author Name(s): Ying Lu,Benchang Zheng, Jun Li,Zongyang Ma,Hang chen. Zhenya Wang,Jinlei Ren,Jia Zhang, Huang Hu. Affiliation(s): XiaoYi AI Lab Nationality(s): China Occupation(s): researcher (These will be listed on the competition website) Bot URL: Personal URL: Affiliation URL: Questions about your bot (please answer as many as you can, especially Q 1-3) Q: What is the overall strategy/strategies of your bot? Why did you choose them? Unlike SAIDA, which only attacks when it’s supplies outstrip it’s component’s, Xiao Yi analyzes the enemy’s initial buildings or units, and decide which strategies to use. The strategies contain fast-attack, Dropship-attack or defence-first, etc. Q: Did you incorporate any of the following AI techniques in your bot? We used rule-based strategies based on SAIDA’s original code. Becide the new strategies writed above, We also add new units based on Hard-coded(state machine),like Valkyrie, Medics, and make Marine active(Which only defence in original SAIDA) Q: How did you become interested in Starcraft AI? The success of AlphaGo SAIDA,and CSE. Q: How long have you been working on your bot? 3 months. Q: About how many lines of code is your bot? Q: Why did you choose the race of your bot? Due to the success of SAIDA last year, We think Terran is challenging and have much to improve. Also, there are few people develop Terran AI bot. Some of our team members are Starcraft Terran players. Q: Did you use any existing code as the basis for your bot? If so, why, and what did you change? SAIDA, as writed above. Q: What do you feel are the strongest and weakest parts of your bot's overall performance? The strongest part of XiaoYi is fast attack,and fast research and upgrade.I think most people ignore the importance of upgrade. The weakest part is rale strategies which seldom appear in starcraft AI competition,like the use of High Templar. Q: If you competed in previous tournaments, what did you change for this year's entry? Q: Have you tested your bot against humans? If so, how did it go? Yes. XiaoYi performs well when it counter Protoss or Terran Players, Not so good with Zerg. 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? Wargame. Its another fantastic platform to test AI technique. 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? I believe there is still a long way before computers can beat humans. We have done a lot at macro-management, but micro-management still have a long way to go. If used well, several marines can kill dozens of Zerglings. Q: What do you feel is the biggest hurdle (technological or otherwise) in improving your bot's AI? SAIDA is rule-based, so there is probably the biggest hurdle to combine AI techniques to completely rules. We didn’t have enough time to complete this year, and we will definitely made this came true next year. Q: Which bots are the most interesting to you and why? CSE DaQin and SAIDA from AIIDE 2018 competition. CSE and DaQin are Chinese teams,we are interested in how they have improved the original Locutus bot. CSE uses MLP to predict build orders, that’s amazing. On the other hand, SAIDA is rule-based, and simulate as much game state as possible, That is awesome. AIIDE Specific Question: Q: Do you feel that the current format of iterated round-robin win percentage is a good indicator of bot skill ranking? If not, how would you change it?