# Noisy Networks for Exploration

 Authors Meire Fortunato, Bilal Piot, Olivier Pietquin, Ian Osband, Vlad Mnih, Alex Graves, Demis Hassabis, Shane Legg, Remi Munos, Charles Blundell, Jacob Menick, Mohammad Gheshlaghi Azar Journal/Conference Name ICLR 2018 1 Paper Category Artificial Intelligence Paper Abstract We introduce NoisyNet, a deep reinforcement learning agent with parametric noise added to its weights, and show that the induced stochasticity of the agent's policy can be used to aid efficient exploration. The parameters of the noise are learned with gradient descent along with the remaining network weights. NoisyNet is straightforward to implement and adds little computational overhead. We find that replacing the conventional exploration heuristics for A3C, DQN and dueling agents (entropy reward and $\epsilon$-greedy respectively) with NoisyNet yields substantially higher scores for a wide range of Atari games, in some cases advancing the agent from sub to super-human performance. Date of publication 2017 Code Programming Language Multiple Comment