Loading subject…
Cumulative reward
The sum of every reward a reinforcement learning agent collects across a complete episode, which is the quantity the agent tries to maximise.
Discount factor
A value between 0 and 1, written $\gamma$, that sets how much a later reward is worth to a reinforcement learning agent now. A reward $k$ steps ahead counts as $\gamma^k$ times its face value.
Epsilon-greedy
A strategy that takes a random action with probability $\epsilon$ and the highest-valued known action the rest of the time, giving the agent a set rate of exploration.
Exploration and exploitation trade-off
The choice a reinforcement learning agent faces at every step between taking the action it currently believes is best and trying an action whose value it is still uncertain about.
Policy (reinforcement learning)
The agent's rule for choosing what to do: a mapping from each state of the environment to the action the agent will take in that state.
Q-learning
A reinforcement learning method that stores an estimated value for every state and action pair and updates it after each step, using the reward received and the best estimate available in the next state.
Reinforcement learning
An agent takes actions in an environment and receives a reward or a penalty in return.