Hierarchal Decomposition of Neural Data using Boosted Mixtures of Hidden Markov Chains and its application to a BMI

In this paper, we propose a simple algorithm that takes multidimensional neural input data and decomposes the joint likelihood into marginals using boosted mixtures of hidden Markov chains (BM-HMM). The algorithm applies techniques from boosting to create hierarchal dependencies between these marginal subspaces. Finally, borrowing ideas from mixture of experts, the local information is weighted and incorporated into an ensemble decision. Our results show that this algorithm is very simple to train and computationally efficient, while also providing the ability to reduce the input dimensionality for brain machine interfaces (BMIs).

[1]  Michael I. Jordan,et al.  Hidden Markov Decision Trees , 1996, NIPS.

[2]  Dawn M. Taylor,et al.  Extraction algorithms for cortical control of arm prosthetics , 2001, Current Opinion in Neurobiology.

[3]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[4]  Jerald D. Kralik,et al.  Chronic, multisite, multielectrode recordings in macaque monkeys , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[5]  G. Schöner,et al.  Preshaping and continuous evolution of motor cortical representations during movement preparation , 2003, The European journal of neuroscience.

[6]  R. Jindra.  Mass action in the nervous system W. J. Freeman, Academic Press, New York (1975), 489 pp., (hard covers). $34.50 , 1976, Neuroscience.

[7]  Nathan Intrator,et al.  Boosted Mixture of Experts: An Ensemble Learning Scheme , 1999, Neural Computation.

[8]  Rajesh P. N. Rao.  Bayesian Computation in Recurrent Neural Circuits , 2004, Neural Computation.

[9]  Alex Acero,et al.  Spoken Language Processing: A Guide to Theory, Algorithm and System Development , 2001 .

[10]  José Carlos Príncipe,et al.  Bimodal brain-machine interface for motor control of robotic prosthetic , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[11]  Jerald D. Kralik,et al.  Real-time prediction of hand trajectory by ensembles of cortical neurons in primates , 2000, Nature.

[12]  Michael I. Jordan,et al.  Factorial Hidden Markov Models , 1995, Machine Learning.

[13]  Emanuel Todorov,et al.  On the role of primary motor cortex in arm movement control , 2003 .

[14]  Deniz Erdogmus,et al.  Divide-and-conquer approach for brain machine interfaces: nonlinear mixture of competitive linear models , 2003, Neural Networks.

[15]  Yoav Freund,et al.  Boosting the margin: A new explanation for the effectiveness of voting methods , 1997, ICML.

[16]  Michael I. Jordan,et al.  Learning Fine Motion by Markov Mixtures of Experts , 1995, NIPS.

[17]  Donald O. Walter,et al.  Mass action in the nervous system , 1975 .

[18]  Robert E. Schapire,et al.  The strength of weak learnability , 1990, Mach. Learn..

[19]  Geoffrey E. Hinton,et al.  Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[20]  Manel Martínez-Ramón,et al.  fMRI pattern classification using neuroanatomically constrained boosting , 2006, NeuroImage.

[21]  Jiri Matas,et al.  On Combining Classifiers , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[22]  Deniz Erdogmus,et al.  Input-output mapping performance of linear and nonlinear models for estimating hand trajectories from cortical neuronal firing patterns , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[23]  Jian Li,et al.  Iterative RELIEF for feature weighting , 2006, ICML.

[24]  Matthew Brand,et al.  Coupled hidden Markov models for modeling interacting processes , 1997 .

[25]  L. Baum,et al.  A Maximization Technique Occurring in the Statistical Analysis of Probabilistic Functions of Markov Chains , 1970 .