Inductive Pattern Formation
MetadataShow full item record
With the extended computational limits of algorithmic recursion, scientific investigation is transitioning away from computationally decidable problems and beginning to address computationally undecidable complexity. The analysis of deductive inference in structure-property models are yielding to the synthesis of inductive inference in process-structure simulations. Process-structure modeling has examined external order parameters of inductive pattern formation, but investigation of the internal order parameters of self-organization have been hampered by the lack of a mathematical formalism with the ability to quantitatively define a specific configuration of points. This investigation addressed this issue of quantitative synthesis. Local space was developed by the Poincare inflation of a set of points to construct neighborhood intersections, defining topological distance and introducing situated Boolean topology as a local replacement for point-set topology. Parallel development of the local semi-metric topological space, the local semi-metric probability space, and the local metric space of a set of points provides a triangulation of connectivity measures to define the quantitative architectural identity of a configuration and structure independent axes of a structural configuration space. The recursive sequence of intersections constructs a probabilistic discrete spacetime model of interacting fields to define the internal order parameters of self-organization, with order parameters external to the configuration modeled by adjusting the morphological parameters of individual neighborhoods and the interplay of excitatory and inhibitory point sets. The evolutionary trajectory of a configuration maps the development of specific hierarchical structure that is emergent from a specific set of initial conditions, with nested boundaries signaling the nonlinear properties of local causative configurations. This exploration of architectural configuration space concluded with initial process-structure-property models of deductive and inductive inference spaces. In the computationally undecidable problem of human niche construction, an adaptive-inductive pattern formation model with predictive control organized the bipartite recursion between an information structure and its physical expression as hierarchical ensembles of artificial neural network-like structures. The union of architectural identity and bipartite recursion generates a predictive structural model of an evolutionary design process, offering an alternative to the limitations of cognitive descriptive modeling. The low computational complexity of these models enable them to be embedded in physical constructions to create the artificial life forms of a real-time autonomously adaptive human habitat.
architectural configuration space
Boolean neighborhood group
completely connected topological space
completely developed space
deductive inference graph
deductive inference space
human niche construction
in absentia modeling
inductive inference space
internal order parameter
local phase transition
local probability space
local sample space
locally complete metric space
locally complete probability space
probabilistic fitness landscape
semi-metric probability space
semi-metric topological space
situated conditional probability chain
situated structural modeling
structural phase transition
Mullican, Raymond Charles (2017). Inductive Pattern Formation. Doctoral dissertation, Texas A & M University. Available electronically from
Showing items related by title, author, creator and subject.
A finite element model for transient thermal/structural analysis of large composite space structures Lutz, James Delmar (Texas A&M University, 1986)Not available
Local structure and thermodynamic properties of model fluids containing rod-like polymers: a comparison of theory and Monte Carlo simulation Dane, D. Merrill (Texas A&M University, 1997)Rigid rod polymers are an interesting and commercially important class of materials. To manipulate these materials on the macroscopic scale, one must understand their behavior on the molecular level. This research investigated ...
Choosing good distance metrics and local planners for probabilistic roadmap motion planning methods Bayazit, Osman Burchan (Texas A&M University, 1998)This thesis presents a comparative evaluation of different distance metrics and local planners within the context of probabilistic roadmap methods for motion planning. Both C-space and Workspace distance metrics and local ...