Header logo is


2005


no image
Composite adaptive control with locally weighted statistical learning

Nakanishi, J., Farrell, J. A., Schaal, S.

Neural Networks, 18(1):71-90, January 2005, clmc (article)

Abstract
This paper introduces a provably stable learning adaptive control framework with statistical learning. The proposed algorithm employs nonlinear function approximation with automatic growth of the learning network according to the nonlinearities and the working domain of the control system. The unknown function in the dynamical system is approximated by piecewise linear models using a nonparametric regression technique. Local models are allocated as necessary and their parameters are optimized on-line. Inspired by composite adaptive control methods, the proposed learning adaptive control algorithm uses both the tracking error and the estimation error to update the parameters. We first discuss statistical learning of nonlinear functions, and motivate our choice of the locally weighted learning framework. Second, we begin with a class of first order SISO systems for theoretical development of our learning adaptive control framework, and present a stability proof including a parameter projection method that is needed to avoid potential singularities during adaptation. Then, we generalize our adaptive controller to higher order SISO systems, and discuss further extension to MIMO problems. Finally, we evaluate our theoretical control framework in numerical simulations to illustrate the effectiveness of the proposed learning adaptive controller for rapid convergence and high accuracy of control.

am

link (url) [BibTex]

2005


link (url) [BibTex]


no image
Global phase diagrams of binary dipolar fluid mixtures

Szalai, I., Dietrich, S.

Molecular Physics, 103(21-23):2873-2895, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Reversible diffusion-limited reactions: "Chemical Equilibrium" state and the Law of Mass Action revisited

Voituriez, R., Moreau, M., Oshanin, G.

Europhysics Letters, 69(2):177-183, 2005 (article)

icm

[BibTex]

[BibTex]


no image
The depletion potential in one, two and three dimensions

Roth, R., König, P. M.

Pramana - Journal of Physics, 64(6):971-980, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Saltatory drift in a randomly driven two-wave potential

Oshanin, G., Klafter, J., Urbakh, M.

Journal of Physics: Condensed Matter, 17(47, Sp. Iss. SI):S3697-S3707, 2005 (article)

icm

[BibTex]

[BibTex]


no image
A thin-film equation for viscoelastic liquids of Jeffreys type

Rauscher, M., Münch, A., Wagner, B., Blossey, R.

European Physical Journal E, 17(3):373-379, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Curvature expansion of density profiles

König, P.-M., Bryk, P., Mecke, K. R., Roth, R.

Europhysics Letters, 69(5):832-838, 2005 (article)

icm

[BibTex]

[BibTex]


no image
General methods for free-volume theory

Oversteegen, S. M., Roth, R.

Journal of Chemical Physics, 122(21), 2005 (article)

icm

[BibTex]

[BibTex]


no image
Morphological characterisation of point patterns

Mecke, K., Stoyan, D.

Biometrical Journal, 47, pages: 1-16, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Thermodynamic Casimir force in models of 4He films

Dantchev, D., Krech, M., Dietrich, S.

Physical Review Letters, 95(25), 2005 (article)

icm

[BibTex]

[BibTex]


no image
Capillary-induced interactions between colloids at an interface

Dominguez, A., Oettel, M., Dietrich, S.

Journal of Physics: Condensed Matter, 17(45):S3387-S3392, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Homogeneous and inhomogeneous hard-sphere mixtures: manifestations of structural crossover

Grodon, C., Dijkstra, M., Evans, R., Roth, R.

Molecular Physics, 103, pages: 3009-3023, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Critical behavior of the two-dimensional randomly driven lattice gas

Caracciolo, S., Gambassi, A., Gubinelli, M., Pelissetto, A.

Physical Review E, 72(5), 2005 (article)

icm

[BibTex]

[BibTex]


no image
Integral equations for simple fluids in a general reference functional approach

Oettel, M.

Journal of Physics: Condensed Matter, 17(3):429-452, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Ageing properties of critical systems

Calabrese, P., Gambassi, A.

Journal of Physics A, 38(18):R133-R193, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Physics of size selectivity

Roth, R., Gillespie, D.

Physical Review Letters, 95, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Solvent mediated interactions close to fluid-fluid phase separation: microscopic treatment of bridging in a soft-core fluid

Archer, A. J., Evans, R., Roth, R., Oettel, M.

Journal of Chemical Physics, 122, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Phase behavior of a nematic liquid crystal in contact with a chemically and geometrically structured substrate

Harnau, L., Kondrat, S., Poniewierski, A.

Physical Review E, 72, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Natural Actor-Critic

Peters, J., Vijayakumar, S., Schaal, S.

In Proceedings of the 16th European Conference on Machine Learning, 3720, pages: 280-291, (Editors: Gama, J.;Camacho, R.;Brazdil, P.;Jorge, A.;Torgo, L.), Springer, ECML, 2005, clmc (inproceedings)

Abstract
This paper investigates a novel model-free reinforcement learning architecture, the Natural Actor-Critic. The actor updates are based on stochastic policy gradients employing AmariÕs natural gradient approach, while the critic obtains both the natural policy gradient and additional parameters of a value function simultaneously by linear regres- sion. We show that actor improvements with natural policy gradients are particularly appealing as these are independent of coordinate frame of the chosen policy representation, and can be estimated more efficiently than regular policy gradients. The critic makes use of a special basis function parameterization motivated by the policy-gradient compatible function approximation. We show that several well-known reinforcement learning methods such as the original Actor-Critic and BradtkeÕs Linear Quadratic Q-Learning are in fact Natural Actor-Critic algorithms. Em- pirical evaluations illustrate the effectiveness of our techniques in com- parison to previous methods, and also demonstrate their applicability for learning control on an anthropomorphic robot arm.

am ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Diffusive spreading and mixing of fluid monolayers

Popescu, M. N., Dietrich, S., Oshanin, G.

Journal of Physics: Condensed Matter, 17(49):S4189-S4198, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Free isotropic-nematic interfaces in fluids of charged platelike colloids

Bier, M., Harnau, L., Dietrich, S.

Journal of Chemical Physics, 123, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Critical adsorption at chemically structured substrates

Sprenger, M., Schlesener, F., Dietrich, S.

Physical Review E, 71, 2005 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Local density profiles in thin films and multilayers from diffuse x-ray and neutron scattering

Rauscher, M., Reichert, H., Engemann, S., Dosch, H.

Physical Review B, 72(20), 2005 (article)

icm

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Corrections to the law of mass action and properties of the asymptotic t\textequalsinfinity state for reversible diffusion-limited reactions

Voituriez, R., Moreau, M., Oshanin, G.

Journal of Chemical Physics, 122(8), 2005 (article)

icm

[BibTex]


no image
Corrections to local scale invariance in the nonequilibrium dynamics of critical systems: numerical evidences

Pleimling, M., Gambassi, A.

Physical Review B, 71(18), 2005 (article)

icm

[BibTex]


no image
Kinetics of diffusion-limited catalytically activated reactions: an extension of the Wilemski-Fixman approach

Benichou, O., Coppey, M., Moreau, M., Oshanin, G.

The Journal of Chemical Physics, 123, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Adhesive gravitational clustering

Buchert, T., Dominguez, A.

Astronomy & Astrophysics, 438(2):443-460, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Comparative experiments on task space control with redundancy resolution

Nakanishi, J., Cory, R., Mistry, M., Peters, J., Schaal, S.

In Proceedings of the 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages: 3901-3908, Edmonton, Alberta, Canada, Aug. 2-6, IROS, 2005, clmc (inproceedings)

Abstract
Understanding the principles of motor coordination with redundant degrees of freedom still remains a challenging problem, particularly for new research in highly redundant robots like humanoids. Even after more than a decade of research, task space control with redundacy resolution still remains an incompletely understood theoretical topic, and also lacks a larger body of thorough experimental investigation on complex robotic systems. This paper presents our first steps towards the development of a working redundancy resolution algorithm which is robust against modeling errors and unforeseen disturbances arising from contact forces. To gain a better understanding of the pros and cons of different approaches to redundancy resolution, we focus on a comparative empirical evaluation. First, we review several redundancy resolution schemes at the velocity, acceleration and torque levels presented in the literature in a common notational framework and also introduce some new variants of these previous approaches. Second, we present experimental comparisons of these approaches on a seven-degree-of-freedom anthropomorphic robot arm. Surprisingly, one of our simplest algorithms empirically demonstrates the best performance, despite, from a theoretical point, the algorithm does not share the same beauty as some of the other methods. Finally, we discuss practical properties of these control algorithms, particularly in light of inevitable modeling errors of the robot dynamics.

am ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Effective pair interactions between colloidal particles at a nematic-isotropic interface

Andrienko, D., Tasinkevych, M., Dietrich, S.

Europhysics Letters, 70(1):95-101, 2005 (article)

icm

[BibTex]

[BibTex]


no image
The Boolean Model: from Matheron till today

Stoyan, D., Mecke, K.

In Space, Structure and Randomness: contributions in honor of Georges Matheron in the fields of geostatistics, random sets, and mathematical morphology, 183, pages: 151-182, Lecture Notes in Statistics, Springer, New York, 2005 (incollection)

icm

[BibTex]

[BibTex]


no image
On thermal fluctuations in thin film flow

Mecke, K., Rauscher, M.

Journal of Physics: Condensed Matter, 17(45):S3515-S3522, 2005 (article)

icm

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Scaling in the time domain: Universal dynamics of order fluctuations in Fe3Al

Mocuta, C., Reichert, H., Mecke, K., Dosch, H., Drakopoulos, M.

Science, 308(5726):1287-1291, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Structure and equation of state of interaction site models for disc-shaped lamellar colloids

Costa, D., Hansen, J.-P., Harnau, L.

Molecular Physics, 103(14):1917-1927, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Local density profiles in thin films and multilayers from diffuse x-ray and neutron scattering

Rauscher, M., Reichert, H., Engemann, S., Dosch, H.

Physical Review B, 72, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Fluids in porous media: a morphometric approach

Mecke, K., Arns, C. H.

Journal of Physics: Condensed Matter, 17, pages: S503-S534, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Fluid mixtures at curved walls

Roth, R.

Journal of Physics: Condensed Matter, 17, pages: S3463-S3468, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Microscopic model of charge carrier transfer in complex media

Benichou, O., Klafter, J., Moreau, M., Oshanin, G.

Chemical Physics, 319(1-3):16-27, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Wetting on structured substrates

Dietrich, S., Popescu, M. N., Rauscher, M.

Journal of Physics: Condensed Matter, 17(9):S577-S593, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Attractions between charged colloids at water interfaces

Oettel, M., Dominguez, A., Dietrich, S.

Journal of Physics: Condensed Matter, 17(32):L337-L342, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Effective capillary interaction of spherical particles at fluid interfaces

Oettel, M., Dominguez, A., Dietrich, S.

Physical Review E, 71, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Local orientations of fluctuating fluid interfaces

Mecke, K., Dietrich, S.

Journal of Chemical Physics, 123(20), 2005 (article)

icm

[BibTex]

[BibTex]


no image
Evidence for capillarity contributions to gecko adhesion from single spatula nanomechanical measurements

Huber, G., Mantz, H., Spolenak, R., Mecke, K., Jacobs, K., Gorb, S. N., Arzt, E.

Proceedings of the National Academy of Sciences of the United States, 102(45):16293-16296, 2005 (article)

icm

[BibTex]

[BibTex]


no image
Swollen lamellar phases confined in capillary-like pores

Tasinkevych, M., Ciach, A.

Physical Review E, 72(6), 2005 (article)

icm

[BibTex]

[BibTex]


no image
Nematic liquid crystal in contact with periodically patterned surfaces

Kondrat, S., Poniewierski, A., Harnau, L.

Liquid Crystals, 32(1):95-105, 2005 (article)

icm

[BibTex]

[BibTex]


no image
A model of smooth pursuit based on learning of the target dynamics using only retinal signals

Shibata, T., Tabata, H., Schaal, S., Kawato, M.

Neural Networks, 18, pages: 213-225, 2005, clmc (article)

Abstract
While the predictive nature of the primate smooth pursuit system has been evident through several behavioural and neurophysiological experiments, few models have attempted to explain these results comprehensively. The model we propose in this paper in line with previous models employing optimal control theory; however, we hypothesize two new issues: (1) the medical superior temporal (MST) area in the cerebral cortex implements a recurrent neural network (RNN) in order to predict the current or future target velocity, and (2) a forward model of the target motion is acquired by on-line learning. We use stimulation studies to demonstrate how our new model supports these hypotheses.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Predicting EMG Data from M1 Neurons with Variational Bayesian Least Squares

Ting, J., D’Souza, A., Yamamoto, K., Yoshioka, T., Hoffman, D., Kakei, S., Sergio, L., Kalaska, J., Kawato, M., Strick, P., Schaal, S.

In Advances in Neural Information Processing Systems 18 (NIPS 2005), (Editors: Weiss, Y.;Schölkopf, B.;Platt, J.), Cambridge, MA: MIT Press, Vancouver, BC, Dec. 6-11, 2005, clmc (inproceedings)

Abstract
An increasing number of projects in neuroscience requires the statistical analysis of high dimensional data sets, as, for instance, in predicting behavior from neural firing, or in operating artificial devices from brain recordings in brain-machine interfaces. Linear analysis techniques remain prevalent in such cases, but classi-cal linear regression approaches are often numercially too fragile in high dimen-sions. In this paper, we address the question of whether EMG data collected from arm movements of monkeys can be faithfully reconstructed with linear ap-proaches from neural activity in primary motor cortex (M1). To achieve robust data analysis, we develop a full Bayesian approach to linear regression that automatically detects and excludes irrelevant features in the data, and regular-izes against overfitting. In comparison with ordinary least squares, stepwise re-gression, partial least squares, and a brute force combinatorial search for the most predictive input features in the data, we demonstrate that the new Bayesian method offers a superior mixture of characteristics in terms of regularization against overfitting, computational efficiency, and ease of use, demonstrating its potential as a drop-in replacement for other linear regression techniques. As neuroscientific results, our analyses demonstrate that EMG data can be well pre-dicted from M1 neurons, further opening the path for possible real-time inter-faces between brains and machines.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Rapbid synchronization and accurate phase-locking of rhythmic motor primitives

Pongas, D., Billard, A., Schaal, S.

In IEEE International Conference on Intelligent Robots and Systems (IROS 2005), pages: 2911-2916, Edmonton, Alberta, Canada, Aug. 2-6, 2005, clmc (inproceedings)

Abstract
Rhythmic movement is ubiquitous in human and animal behavior, e.g., as in locomotion, dancing, swimming, chewing, scratching, music playing, etc. A particular feature of rhythmic movement in biology is the rapid synchronization and phase locking with other rhythmic events in the environment, for instance music or visual stimuli as in ball juggling. In traditional oscillator theories to rhythmic movement generation, synchronization with another signal is relatively slow, and it is not easy to achieve accurate phase locking with a particular feature of the driving stimulus. Using a recently developed framework of dynamic motor primitives, we demonstrate a novel algorithm for very rapid synchronizaton of a rhythmic movement pattern, which can phase lock any feature of the movement to any particulur event in the driving stimulus. As an example application, we demonstrate how an anthropomorphic robot can use imitation learning to acquire a complex rumming pattern and keep it synchronized with an external rhythm generator that changes its frequency over time.

am

link (url) [BibTex]

link (url) [BibTex]