It's stated in the article. During a day's play, the AI suggests plays based on its current knowledge. At night, the day's events are fed into the system for it

It's stated in the article. During a day's play, the AI suggests plays based on its current knowledge. At night, the day's events are fed into the system for it to learn on for the next day. I assume there are probably papers that specify what form of learning is taking place, but the article didn't go into that level of detail and I haven't tried to track it down.