Dueling Network Architectures for Deep Reinforcement Learning外文学习材料.pdfVIP

Dueling Network Architectures for Deep Reinforcement Learning Ziyu Wang ZIYU @GOOGLE .COM Tom Schaul SCHAUL @GOOGLE .COM Matteo Hessel MTTHSS @GOOGLE .COM Hado van Hasselt HADO @GOOGLE .COM Marc Lanctot LANCTOT @GOOGLE .COM Nando de Freitas NANDODEFREITAS @GMAIL .COM 6 Google DeepMind, London, UK 1 0 2 r Abstract In spite of this, most of the approaches for RL use standard p neural networks, such as convolutional networks, MLPs, A In recent years there have been many successes LSTMs and autoencoders. The focus in these recent ad- of using deep representations in reinforcement vances has been on designing improved control and RL al- 5 learning. Still, many of these applications use gorithms, or simply on incorporating existing neural net- ] conventional architectures, such as convolutional work architectures into RL methods. Here, we take an al- G networks, LSTMs, or auto-encoders. In this pa- ternative but complementary approach of focusing primar- L per, we present a new neural network architec- ily on innovating a neural network architecture that is better s. ture for model-free reinforcement learning. Our