NoisyNet-A3C

Introduced by Fortunato et al. in Noisy Networks for Exploration

NoisyNet-A3C is a modification of A3C that utilises noisy linear layers for exploration instead of $\epsilon$-greedy exploration as in the original DQN formulation.

Source: Noisy Networks for Exploration

Read Paper See Code

Papers

Paper	Code	Results	Date	Stars

Tasks

Task	Papers	Share
Atari Games	1	33.33%
Efficient Exploration	1	33.33%
Reinforcement Learning (RL)	1	33.33%

Usage Over Time

This feature is experimental; we are continuously improving our matching algorithm.

Components

Component	Type	Add Remove
A3C	Policy Gradient Methods
Noisy Linear Layer	Randomized Value Functions

Categories

Add Remove

Policy Gradient Methods