Free Information Center
Empowerment in machine learning is an information-theoretic measure used to quantify the agent’s potential influence over its environment. It is applied primarily in reinforcement learning as an intrinsic motivation to encourage exploration and autonomous behavior.
Exploration by random network distillation (RND) is a technique in reinforcement learning used to encourage agents to explore unfamiliar states by providing intrinsic rewards based on the novelty of observations. It leverages a fixed random neural network and a trainable network to measure prediction error as a proxy for novelty.
Curiosity-driven reinforcement learning is a subfield of reinforcement learning that incorporates intrinsic motivation mechanisms, encouraging agents to explore their environment beyond external rewards. This approach aims to improve learning efficiency in complex or sparse reward settings by fostering exploration through curiosity.