2014/12/29 by James P. Crutchfield, Ryan G. James, Crutchfield, James P. +5
Biochemistry, Genetics and Molecular Biology · Computer Science · Neuroscience · #Artificial Intelligence (cs.AI) #Chaotic Dynamics (nlin.CD) #Computational Engineering #FOS: Computer and information sciences #FOS: Physical sciences #Finance #Gaussian Processes and Bayesian Inference #Gene Regulatory Network Analysis #Information Theory (cs.IT) #Neural Networks and Applications #Neural dynamics and brain function #Statistical Mechanics (cond-mat.stat-mech) #and Science (cs.CE)
paper · pdf · doi:10.48550/arxiv.1412.8520
openalex publication_date 2014/12/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
We recount recent history behind building compact models of nonlinear,\ncomplex processes and identifying their relevant macroscopic patterns or\n"macrostates". We give a synopsis of computational mechanics, predictive\nrate-distortion theory, and the role of information measures in monitoring\nmodel complexity and predictive performance. Computational mechanics provides a\nmethod to extract the optimal minimal predictive model for a given process.\nRate-distortion theory provides methods for systematically approximating such\nmodels. We end by commenting on future prospects for developing a general\nframework that automatically discovers optimal compact models. As a response to\nthe manuscript cited in the title above, this brief commentary corrects\npotentially misleading claims about its state space compression method and\nplaces it in a broader historical setting.\n