Influence-Based Model Decomposition

Recent rapid advances in MEMS and information processing technology have enabled a new generation of AI robotic systems -- so-called Smart Matter systems - that are sensor rich and physically embedded. These systems range from decentralized control systems that regulate building temperature (smart buildings) to vehicle on-board diagnostic and control systems that interrogate large amounts of sensor data. One of the core tasks in the construction and operation of these Smart Matter systems is to synthesize optimal control policies using data rich models for the systems and environment. Unfortunately, these models may contain thousands of coupled real-valued variables and are prohibitively expensive to reason about using traditional optimization techniques such as neural nets and genetic algorithms. This paper introduces a general mechanism for automatically decomposing a large model into smaller subparts so that these subparts can be separately optimized and then combined. The mechanism decomposes a model using an influence graph that records the coupling strengths among constituents of the model. This paper demonstrates the mechanism in an application of decentralized optimization for a temperature regulation problem. Performance data has shown that the approach is much more efficient than the standard discrete optimization algorithms and achieves comparable accuracy.

[1]  T. Chan,et al.  Domain decomposition algorithms , 1994, Acta Numerica.

[2]  Anoop K. Dhingra,et al.  OPTIMAL PLACEMENT OF ACTUATORS IN ACTIVELY CONTROLLED STRUCTURES , 1994 .

[3]  Avi Pfeffer,et al.  Structured Representation of Complex Stochastic Systems , 1998, AAAI/IAAI.

[4]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[5]  Feng Zhao.  An O(N) Algorithm for Three-dimensional N-body Simulations , 2022 .

[6]  P. Pandurang Nayak,et al.  Immobile Robots AI in the New Millennium , 1996, AI Mag..

[7]  Gun-Shing Chen,et al.  OPTIMAL PLACEMENT OF ACTIVE/PASSIVE MEMBERS IN TRUSS STRUCTURES USING SIMULATED ANNEALING , 1991 .

[8]  Jitendra Malik,et al.  Normalized cuts and image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[9]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[10]  Chris Bailey-Kellogg,et al.  Qualitative Analysis of Distributed Physical Systems with Applications to Control Synthesis , 1998, AAAI/IAAI.

[11]  Bill Millar,et al.  Automated Decomposition of Model-based Learning Problems , 1996 .

[12]  Benjamin Kuipers,et al.  Model Decomposition and Simulation: A Component Based Qualitative Simulation Algorithm , 1997, AAAI/IAAI.

[13]  Horst D. Simon,et al.  Partitioning of unstructured problems for parallel processing , 1991 .

[14]  Feng Zhao,et al.  An {\it bf O(N)} Algorithm for Three-Dimensional N-body Simulations , 1987 .

[15]  Charalabos C. Doumanidis,et al.  In-process control in thermal rapid prototyping , 1997 .

[16]  Daniel P. Huttenlocher,et al.  Image segmentation using local variation , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).