3 ms·
Please call me again when they actually solve the exploration problem instead of falling back to a good example. People beating this game do not do it based on
by AstralStorm 8y ago
Please call me again when they actually solve the exploration problem instead of falling back to a good example.
People beating this game do not do it based on a let's play video.
- red75prime 8y agoAnd no human ever was able to play montezuma's revenge, given only a let's play video. Usually it requires 5-10 years of accumulating prior information. It is still remarkable achievement.
- alexcnwy 8y agoThis is highly significant because it was one of the games that the Atari games that the original Deep Q learning model could not beat.
- backpropaganda 8y agoDQN is solving a model-free RL problem. This method is not. You're not allowed to reset states in model-free RL. If you have access to the model/simulator which allows you to reset, you might as well use a model-based method like MCTS.