Information Maximization (InfoMax)
Definition
A principle for self-supervised learning stating that representations should maximize mutual information between the input and the representation, ensuring representations retain as much useful signal from the input as possible. Formalized by Bell and Sejnowski in 1995, it became foundational for many self-supervised methods and is a key idea influencing JEPA approaches.
Cast within
No topic-region cast yet — this appears once Information Maximization (InfoMax)'s compiled claims are aligned into a topic region's argument tree.
Claims mentioning Information Maximization (InfoMax) (1)
Your notes
Loading notes...