Evolutionary Hessian Learning: Forced Optimal Covariance Adaptive Learning (FOCAL)

The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) has been the most successful Evolution Strategy at exploiting covariance information; it uses a form of Principle Component Analysis which, under certain conditions, is suggested to converge to the correct covariance matrix, formulated as the inverse of the mathematically well-defined Hessian matrix. However, in practice, there exist conditions where CMA-ES converges to the global optimum (accomplishing its primary goal) while it does not learn the true covariance matrix (missing an auxiliary objective), likely due to step-size deficiency. These circumstances can involve high-dimensional landscapes with large condition numbers. This paper introduces a novel technique entitled Forced Optimal Covariance Adaptive Learning (FOCAL), with the explicit goal of determining the Hessian at the global basin of attraction. It begins by introducing theoretical foundations to the inverse relationship between the learned covariance and the Hessian matrices. FOCAL is then introduced and demonstrated to retrieve the Hessian matrix with high fidelity on both model landscapes and experimental Quantum Control systems, which are observed to possess a non-separable, non-quadratic search landscape. The recovered Hessian forms are corroborated by physical knowledge of the systems. This study constitutes an example for Natural Computing successfully serving other branches of natural sciences, and introducing at the same time a powerful generic method for any high-dimensional continuous search seeking landscape information.

[1]  Herschel Rabitz,et al.  Laboratory observation of quantum control level sets , 2006 .

[2]  Monte Lunacek,et al.  Calibration of liquid crystal ultrafast pulse shaper with common-path spectral interferometry and application to coherent control with a covariance matrix adaptation evolutionary strategy. , 2008, The Review of scientific instruments.

[3]  Hans-Paul Schwefel,et al.  Evolution strategies – A comprehensive introduction , 2002, Natural Computing.

[4]  Ofer M. Shir,et al.  On the diversity of multiple optimal controls for quantum systems , 2008 .

[5]  H. Rabitz,et al.  Teaching lasers to control molecules. , 1992, Physical review letters.

[6]  H. Rabitz,et al.  Why do effective quantum controls appear easy to find , 2006 .

[7]  Keith E. Mathias,et al.  In Parallel Problem Solving from Nature-PPSN III , 1994 .

[8]  Stefan Roth,et al.  Covariance Matrix Adaptation for Multi-objective Optimization , 2007, Evolutionary Computation.

[9]  Nikolaus Hansen,et al.  Adapting arbitrary normal mutation distributions in evolution strategies: the covariance matrix adaptation , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[10]  Thomas Bäck,et al.  Evolution Strategies for Laser Pulse Compression , 2007, Artificial Evolution.

[11]  T. Baumert,et al.  Femtosecond pulse shaping by an evolutionary algorithm with feedback , 1997 .

[12]  P. H. Bucksbaum,et al.  Coherent control using adaptive learning algorithms , 2001 .

[13]  Arnold Neumaier,et al.  Solving Ill-Conditioned and Singular Linear Systems: A Tutorial on Regularization , 1998, SIAM Rev..

[14]  Nikolaus Hansen,et al.  A Derandomized Approach to Self-Adaptation of Evolution Strategies , 1994, Evolutionary Computation.

[15]  Silberberg,et al.  Transform-Limited Pulses Are Not Optimal for Resonant Multiphoton Transitions. , 2001, Physical review letters.

[16]  Ofer M. Shir,et al.  Evolutionary multi-objective quantum control experiments with the covariance matrix adaptation , 2009, GECCO '09.

[17]  Jing J. Liang,et al.  Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-Parameter Optimization , 2005 .

[18]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[19]  Ofer M. Shir,et al.  Adaptive Niche Radii and Niche Shapes Approaches for Niching with the CMA-ES , 2010, Evolutionary Computation.

[20]  H. Rabitz,et al.  Optimal control of quantum-mechanical systems: Existence, numerical approximation, and applications. , 1988, Physical review. A, General physics.

[21]  Ofer M. Shir,et al.  Accelerated optimization and automated discovery with covariance matrix adaptation for experimental quantum control , 2009 .

[22]  Ofer M. Shir,et al.  Evolutionary Algorithms in the Optimization of Dynamic Molecular Alignment , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[23]  Anne Auger,et al.  Performance evaluation of an advanced local search evolutionary algorithm , 2005, 2005 IEEE Congress on Evolutionary Computation.

[24]  Nikolaus Hansen,et al.  Evaluating the CMA Evolution Strategy on Multimodal Test Functions , 2004, PPSN.

[25]  Günter Rudolph,et al.  On Correlated Mutations in Evolution Strategies , 1992, PPSN.

[26]  William H. Press,et al.  Numerical recipes , 1990 .

[27]  Herschel Rabitz,et al.  Quantum control landscapes , 2007, 0710.0684.

[28]  Petros Koumoutsakos,et al.  A Method for Handling Uncertainty in Evolutionary Optimization With an Application to Feedback Control of Combustion , 2009, IEEE Transactions on Evolutionary Computation.

[29]  Kompa,et al.  Whither the future of controlling quantum phenomena? , 2000, Science.

[30]  D. Zeidler,et al.  Evolutionary algorithms and their application to optimal control studies , 2001 .

[31]  Isao Ono,et al.  Functionally specialized CMA-ES: a modification of CMA-ES based on the specialization of the functions of covariance matrix adaptation and step size adaptation , 2008, GECCO '08.

[32]  Nikolaus Hansen,et al.  Step-Size Adaption Based on Non-Local Use of Selection Information , 1994, PPSN.

[33]  Bernhard Sendhoff,et al.  Covariance Matrix Adaptation Revisited - The CMSA Evolution Strategy - , 2008, PPSN.

[34]  David B. Fogel,et al.  Evolutionary algorithms in theory and practice , 1997, Complex.

[35]  Raymond Ros,et al.  A Simple Modification in CMA-ES Achieving Linear Time and Space Complexity , 2008, PPSN.

[36]  Nikolaus Hansen,et al.  On the Adaptation of Arbitrary Normal Mutation Distributions in Evolution Strategies: The Generating Set Adaptation , 1995, ICGA.

[37]  Gustav Gerber,et al.  Femtosecond quantum control of molecular dynamics in the condensed phase. , 2007, Physical chemistry chemical physics : PCCP.

[38]  Ofer M. Shir,et al.  Niching with derandomized evolution strategies in artificial and real-world landscapes , 2009, Natural Computing.

[39]  Ofer M. Shir,et al.  Performance analysis of derandomized evolution strategies in quantum control experiments , 2008, GECCO '08.

[40]  Nikolaus Hansen,et al.  Benchmarking a BI-population CMA-ES on the BBOB-2009 noisy testbed , 2009, GECCO '09.

[41]  H Rabitz,et al.  Coherent control of quantum dynamics: the dream is alive. , 2008, Science.