Reinforcement learning method using two neural networks for stable Q-value estimation.
Machine LearningReinforcement LearningArtificial IntelligenceDeep Learning
Current best complexity is 10:
Double Deep can be discovered in Concepts, categorized under Machine Learning, Reinforcement Learning, Artificial Intelligence, Deep Learning. Reinforcement learning method using two neural networks for stable Q-value estimation.
It has been discovered by 6 boards and crafted 9 times, first appearing 2 months ago. Double Deep can be created from 1 recipe, and goes on to appear as an ingredient in 6 others.
Double Deep was first discovered by Skitsboard 2 months ago. They also hold the lowest-complexity recipe, at 10.
Related concepts worth exploring include Vog, Gemini, Robot, Artificial Intelligence, Whisper, Neural Network, and 6 more.