Building Intelligent Agents Using Deep Reinforcement Learning
no recording on file
summary
Covers using deep reinforcement learning and Monte Carlo Tree Search to build a game-playing agent for the round-based strategy game Hexagon (https://playhexagon.com/). Explains representing game rules, designing the neural network, and the practical steps of training and integrating RL + MCTS to produce a winning agent, with about one-third theory/history and two-thirds implementation details.
AI-written summary · about
related talks
31mDesigning game AI with reinforcement learning
02mDesigning a Video Game AI
56mCan I do AI?
01h 17mTraining intelligent game agents using deep reinforcement learning
01h 03mIEEE CoG 2019 - Industry Day - David Silver (DeepMind)
49mML: Teaching machine to learn from its own mistakes (...)
24mDesigning a Video Game AI with Reinforcement Learning
47mHow we use Reinforcement Learning (...)