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:DaQin Bot Race:Protoss Author Name(s):LionGIS Affiliation(s):liongis@163.com Nationality(s): China Occupation(s):Programmer (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? >>The overall strategy is to attack with ranged units, and if you can attack the enemy, and the enemy can't attack you, the outcome of the game is almost certain. Q: Did you incorporate any of the following AI techniques in your bot? If you did, please be as specific as possible a) Search-Based AI (Path-Finding, A*, MiniMax, MCTS, etc) b) Offline Machine Learning (Supervised or Unsupervised, but not RL) c) Offline Reinforcement Learning d) Online Learning of any kind (Including competition file IO for strategy selection) e) Influence Maps f) Custom Map Analysis g) Hard-coded or rule-based strategy / tactics h) Analysis of bots from previous competitions / hard-coded specific bot counter strategies i) Any techniques not mentioned here >>The current strategy is hard-coded and rule-based, and each confrontation has several options, chosen according to a certain probability. Bot will record the map, opponent ID, results and other information in each game, and select the next strategy with high win rate at the beginning. Q: How did you become interested in Starcraft AI? >>I have been playing starcraft for 20 years and doing program development for 10 years, both of which are my favorite things. I only learned in November 2017 that I can write starcraft AI with apps, and it is the happiest thing for me to combine the two greatest interests, and I will stick to them. Q: How long have you been working on your bot? >>It was known from 2017 that it took only four months to get to the first release, but the main design was a man-to-man fight with humans, and so far, almost a year. Q: About how many lines of code is your bot? >>There are about 40k+ lines of code. Q: Why did you choose the race of your bot? >>I'm a protoss player and know him best, so I chose the protoss. Q: Did you use any existing code as the basis for your bot? If so, why, and what did you change? >>Developing based on Locutus because he is based on UAlbertaBot, which is a good example of what we are talking about. It has a very good framework and thinking. It is also very easy to read. Second, Locutus has been particularly optimized to provide a better foundation for learning AI. However, Locutus does not develop all protoss units, so it loses some initiative and fails to exert all the advantages of the protoss. My Bot tries to make the tactics and arms more diverse and optimizes the algorithm to improve the winning rate of the confrontation. Q: What do you feel are the strongest and weakest parts of your bot's overall performance? >>The most powerful part is to use the characteristics of long range units to attack opponents, such as dragoon, carrier, etc., from a distance to attack enemies and try to avoid being hurt by them. Weak areas are the lack of control over the position of offensive units and formations, resulting in unnecessary loss of many units, strategy and arms production is more rigid, lack of variability. Q: If you competed in previous tournaments, what did you change for this year's entry? >>This is my first time to compete, but I can feel that 2018 is the fastest year for starcraft AI development. This year's opponents are very strong, with more teams and individuals participating, which is a very good thing for starcraft AI development. Q: Have you tested your bot against humans? If so, how did it go? >>I have tried to fight against human beings, and I can beat some Chinese players without knowing the AI strategy, but after a few games, the probability becomes very low. The overall level in China is below the average level. There's a lot of people willing to fight AI. 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? >>I like software development very much, have 10 years of development experience, also like starcraft this game very much, played for nearly 20 years. Starcraft AI is to put my two hobbies together project, I really want to make a contribution in this project, but to develop a truly be able to overcome human AI, has practical application value, and need a lot of manpower and material resources, I would like to join or form a team, if interested can contact me to me: LionGIS@163.com 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? >>At present, starcraft AI still tends to be hard coded and not flexible. Currently, the solution of artificial intelligence is also based on the previous knowledge system, which is not enough to deal with the dynamic nature of real-time strategy games. A new ai system is needed to support and solve unsolved problems. Another limitation is that the current AI is operating on a single machine, which greatly reduces the computing power of AI and needs to improve its computing power. If this new artificial intelligence system comes out and solves the computing power of AI, I think in the next three years or so, AI will definitely win over the most powerful humans. Q: What do you feel is the biggest hurdle (technological or otherwise) in improving your bot's AI? >>The new artificial intelligence system, through the previous knowledge system, to solve the unknown problem. Previous knowledge systems included tactics, habits, psychology, etc. Q: Which bots are the most interesting to you and why? >>The most interesting thing is the CherryPi, because more ai is used in this Bot. Overcoming humans is not our ultimate goal in developing starcraft AI, but it is necessary for these applications to be applied to the real world, to solve something that humans cannot currently solve, and to improve the overall AI capability. 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? >>The round-robin game is supposed to be a fair game for AI right now.