IEEE Deakin student branch is hosting a seminar on “Deep Learning in Games” by Duy Nguyen.
Modern reinforcement learning (RL) is truly marked by the success of deep RL in 2015. A group of researchers from Google’s Deepmind announced a significant breakthrough in their success to eventually create an agent that outperformed a professional player in a series of 49 classic Atari games. In 2016, Google’s Deepmind again created a self-taught AlphaGo program that could beat the best professional Go players, including China’s Ke Jie and Korea’s Lee Sadol. In 2017, OpenAI announced the bot that could beat the best professional gamer on the online game Dota 2 that is supposed to be more complicated than Go. These fates provide the necessary impetus to creating self-playing agents in complicated games. Therefore, in the seminar, we briefly review the state-of-the-art deep RL methods that are widely used in training game-playing agents. We also explain the mechanisms and details in a step-by-step manner. Finally, we introduce a demonstration that involves a self-playing agent in different types of games such as Atari series and Tank Battle City.
Attending is free and everyone is welcome.
Please feel free to register at https://deeplearning_ieeesb.eventbrite.com.au