Ref_Parr

Parr, Da Costa, & Friston (2020) Markov blankets, information geometry and stochastic thermodynamics

gist
マルコフブランケットを持つ力学系において内部状態が統計多様体をパラメータ化し、情報処理と確率的熱力学が直接結合することを証明した論文

Abstract, p. 1:

“In brief, any (weakly mixing) random dynamical system that possesses a Markov blanket—i.e. a separation of internal and external states—is equipped with an information geometry. This means that internal states parametrize a probability density over external states.”
Section 1, p. 2:

“Crucially, this means that there is an explicit link between the inference performed by internal states and their energetics—as characterized by their stochastic thermodynamics.”
Section 3, p. 6:

“More precisely, the partition means that internal states parametrize a probability density over external states. Consequently, the internal state-space has an inherent information geometry (technically, this space is a statistical manifold).”