Jeffreys prior

Jeffreys' rule: The prior distribution for a single parameter [theta] is approximately noninformative if it is taken proportional to the square root of Fisher's information measure.
This rule for the choice of a noninformative prior distribution was first given by Sir Harrold Jeffreys (1961), who justified it on the grounds of its invariance under parameter transformations."
Box and Tiao, 1973