Causal blankets: Theory and algorithmic framework
This provides a theoretical and algorithmic framework for modeling perception-action loops in systems like robotics or AI, though it appears incremental as it builds on existing computational mechanics principles.
The paper tackles the problem of identifying perception-action loops (PALOs) directly from data by introducing a causal blanket framework based on computational mechanics, showing that every bipartite stochastic process has a causal blanket but its effectiveness depends on integrated information.
We introduce a novel framework to identify perception-action loops (PALOs) directly from data based on the principles of computational mechanics. Our approach is based on the notion of causal blanket, which captures sensory and active variables as dynamical sufficient statistics -- i.e. as the "differences that make a difference." Moreover, our theory provides a broadly applicable procedure to construct PALOs that requires neither a steady-state nor Markovian dynamics. Using our theory, we show that every bipartite stochastic process has a causal blanket, but the extent to which this leads to an effective PALO formulation varies depending on the integrated information of the bipartition.