Noisy Networks For Exploration

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.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyfortunato2017noisy

Related papers

• NROWAN-DQN: A Stable Noisy Network With Noise Reduction And Online Weight Adjustment For Exploration (2020)0.00
• Noisy Spiking Actor Network For Exploration (2024)0.00
• Exploring More When It Needs In Deep Reinforcement Learning (2021)0.00
• Parameter Space Noise For Exploration (2017)0.00
• Action Noise In Off-policy Deep Reinforcement Learning: Impact On Exploration And Performance (2022)0.00
• Beyond Noisy-tvs: Noise-robust Exploration Via Learning Progress Monitoring (2025)0.00
• Adaptive Symmetric Reward Noising For Reinforcement Learning (2019)0.00
• Multi-agent Deep Reinforcement Learning With Extremely Noisy Observations (2018)0.00