Curious Agents Saga (Part 1), Legacy of Exploration in Reinforcement Learning

Legacy of Exploration in Reinforcement Learning

Table of Content

• Exploration in Reinforcement Learning

• Classic Explorations

  • ε-greedy

  • Upper Confidence Bound (UCB)

  • Thompson Sampling

  • Information Gain

• Primal Exploration in Deep Reinforcement Learning

  • Entropy Maximization
  • Noisy Networks

Exploration in Reinforcement Learning

In reinforcement learning (RL), an agent interacts with the environment, taking actions a, receiving a reward r, and moving to a new state s. The agent is tasked with maximizing the accumulated rewards or returns R over time by finding optimal actions (policy).

Read the full article