Which parameter is used to gradually reduce epsilon over time, causing exploration to be heavy early and exploitation later?

Which parameter is used to gradually reduce epsilon over time, causing exploration to be heavy early and exploitation later?

• A. Epsilon Greedy Strategy
• B. Q-Learning
• C. Decay Factor
• D. Deep Reinforcement Learning

Answer: (C)

The shift from exploring to exploiting in an epsilon-greedy setup is controlled by a decay factor. In this approach, the agent chooses a random action with probability epsilon and a best-known action with probability 1 - epsilon. To favor exploration early and exploitation later, epsilon is reduced over time using a decay factor, often applied each episode or step: epsilon := max(epsilon_min, epsilon * decay_factor). The decay_factor is a number slightly less than 1, so it gradually lowers epsilon toward a minimum. A value near 1 means slow decay and longer exploration; a smaller value means faster decay and quicker reliance on learned estimates. The other terms describe the strategy itself (epsilon-greedy), the learning algorithm (Q-learning), or the broader field (deep reinforcement learning) rather than the mechanism that reduces epsilon over time.

The shift from exploring to exploiting in an epsilon-greedy setup is controlled by a decay factor. In this approach, the agent chooses a random action with probability epsilon and a best-known action with probability 1 - epsilon. To favor exploration early and exploitation later, epsilon is reduced over time using a decay factor, often applied each episode or step: epsilon := max(epsilon_min, epsilon * decay_factor). The decay_factor is a number slightly less than 1, so it gradually lowers epsilon toward a minimum. A value near 1 means slow decay and longer exploration; a smaller value means faster decay and quicker reliance on learned estimates. The other terms describe the strategy itself (epsilon-greedy), the learning algorithm (Q-learning), or the broader field (deep reinforcement learning) rather than the mechanism that reduces epsilon over time.