Maximum Entropy Principle
Coined by Jaynes, expresses the relation between probability and information.
It states that the probability of some state in a hypothesis space is always entirely conditional on a particular set of knowledge or information.
People who use MEP as a constraint on assigning (prior) probabilities in Bayesianism are usually counted as "objective" Bayesians. This makes sense against the background of Jaynes' effort to link probability theory with information theory: objective Bayesians link the choice of probability distributions in conditionalization to the available background information.
This links the assignment of probabilities to the information theory of entropy because it is at once the acknowledgment of our ignorance: I should maximize my ignorance about the things I have no information about. When we strictly limit ourselves to what we know when assigning prior probabilities (as degrees of belief), we should assume the distribution that maximizes entropy i.e. uncertainty. In that sense, the MEP is a guide for Bayesians on how to choose a prior distribution. The tricky part is, of course, to quantify precisely what we know.
Fiorillo-Beyond_Bayes.md makes a case for a Jaynesian understanding of probabilities and information (in neuroscience). This paper contains an extended and lucid discussion of the maximum entropy principle as coined by Jaynes. In the note, see the sections "Logic vs. information" and "Quantifying the amount of information (and information gain)".
Compare to Laplace's principle of indifference.