Header logo is


2017


On the Design of {LQR} Kernels for Efficient Controller Learning
On the Design of LQR Kernels for Efficient Controller Learning

Marco, A., Hennig, P., Schaal, S., Trimpe, S.

Proceedings of the 56th IEEE Annual Conference on Decision and Control (CDC), pages: 5193-5200, IEEE, IEEE Conference on Decision and Control, December 2017 (conference)

Abstract
Finding optimal feedback controllers for nonlinear dynamic systems from data is hard. Recently, Bayesian optimization (BO) has been proposed as a powerful framework for direct controller tuning from experimental trials. For selecting the next query point and finding the global optimum, BO relies on a probabilistic description of the latent objective function, typically a Gaussian process (GP). As is shown herein, GPs with a common kernel choice can, however, lead to poor learning outcomes on standard quadratic control problems. For a first-order system, we construct two kernels that specifically leverage the structure of the well-known Linear Quadratic Regulator (LQR), yet retain the flexibility of Bayesian nonparametric learning. Simulations of uncertain linear and nonlinear systems demonstrate that the LQR kernels yield superior learning performance.

am ics pn

arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]

2017


arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]


Probabilistic Line Searches for Stochastic Optimization
Probabilistic Line Searches for Stochastic Optimization

Mahsereci, M., Hennig, P.

Journal of Machine Learning Research, 18(119):1-59, November 2017 (article)

pn

link (url) Project Page [BibTex]

link (url) Project Page [BibTex]


Coupling Adaptive Batch Sizes with Learning Rates
Coupling Adaptive Batch Sizes with Learning Rates

Balles, L., Romero, J., Hennig, P.

In Proceedings Conference on Uncertainty in Artificial Intelligence (UAI) 2017, pages: 410-419, (Editors: Gal Elidan and Kristian Kersting), Association for Uncertainty in Artificial Intelligence (AUAI), Conference on Uncertainty in Artificial Intelligence (UAI), August 2017 (inproceedings)

Abstract
Mini-batch stochastic gradient descent and variants thereof have become standard for large-scale empirical risk minimization like the training of neural networks. These methods are usually used with a constant batch size chosen by simple empirical inspection. The batch size significantly influences the behavior of the stochastic optimization algorithm, though, since it determines the variance of the gradient estimates. This variance also changes over the optimization process; when using a constant batch size, stability and convergence is thus often enforced by means of a (manually tuned) decreasing learning rate schedule. We propose a practical method for dynamic batch size adaptation. It estimates the variance of the stochastic gradients and adapts the batch size to decrease the variance proportionally to the value of the objective function, removing the need for the aforementioned learning rate decrease. In contrast to recent related work, our algorithm couples the batch size to the learning rate, directly reflecting the known relationship between the two. On three image classification benchmarks, our batch size adaptation yields faster optimization convergence, while simultaneously simplifying learning rate tuning. A TensorFlow implementation is available.

ps pn

Code link (url) Project Page [BibTex]

Code link (url) Project Page [BibTex]


no image
Dynamic Time-of-Flight

Schober, M., Adam, A., Yair, O., Mazor, S., Nowozin, S.

Proceedings IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017, pages: 170-179, IEEE, Piscataway, NJ, USA, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017 (conference)

ei pn

DOI [BibTex]

DOI [BibTex]


Virtual vs. {R}eal: Trading Off Simulations and Physical Experiments in Reinforcement Learning with {B}ayesian Optimization
Virtual vs. Real: Trading Off Simulations and Physical Experiments in Reinforcement Learning with Bayesian Optimization

Marco, A., Berkenkamp, F., Hennig, P., Schoellig, A. P., Krause, A., Schaal, S., Trimpe, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 1557-1563, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

am ics pn

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]


Fast Bayesian Optimization of Machine Learning Hyperparameters on Large Datasets
Fast Bayesian Optimization of Machine Learning Hyperparameters on Large Datasets

Klein, A., Falkner, S., Bartels, S., Hennig, P., Hutter, F.

Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS 2017), 54, pages: 528-536, Proceedings of Machine Learning Research, (Editors: Sign, Aarti and Zhu, Jerry), PMLR, April 2017 (conference)

pn

pdf link (url) Project Page [BibTex]

pdf link (url) Project Page [BibTex]


no image
Convergence Analysis of Deterministic Kernel-Based Quadrature Rules in Misspecified Settings

Kanagawa, M., Sriperumbudur, B. K., Fukumizu, K.

Arxiv e-prints, arXiv:1709.00147v1 [math.NA], 2017 (article)

Abstract
This paper presents convergence analysis of kernel-based quadrature rules in misspecified settings, focusing on deterministic quadrature in Sobolev spaces. In particular, we deal with misspecified settings where a test integrand is less smooth than a Sobolev RKHS based on which a quadrature rule is constructed. We provide convergence guarantees based on two different assumptions on a quadrature rule: one on quadrature weights, and the other on design points. More precisely, we show that convergence rates can be derived (i) if the sum of absolute weights remains constant (or does not increase quickly), or (ii) if the minimum distance between distance design points does not decrease very quickly. As a consequence of the latter result, we derive a rate of convergence for Bayesian quadrature in misspecified settings. We reveal a condition on design points to make Bayesian quadrature robust to misspecification, and show that, under this condition, it may adaptively achieve the optimal rate of convergence in the Sobolev space of a lesser order (i.e., of the unknown smoothness of a test integrand), under a slightly stronger regularity condition on the integrand.

pn

arXiv [BibTex]

arXiv [BibTex]


Early Stopping Without a Validation Set
Early Stopping Without a Validation Set

Mahsereci, M., Balles, L., Lassner, C., Hennig, P.

arXiv preprint arXiv:1703.09580, 2017 (article)

Abstract
Early stopping is a widely used technique to prevent poor generalization performance when training an over-expressive model by means of gradient-based optimization. To find a good point to halt the optimizer, a common practice is to split the dataset into a training and a smaller validation set to obtain an ongoing estimate of the generalization performance. In this paper we propose a novel early stopping criterion which is based on fast-to-compute, local statistics of the computed gradients and entirely removes the need for a held-out validation set. Our experiments show that this is a viable approach in the setting of least-squares and logistic regression as well as neural networks.

ps pn

link (url) Project Page Project Page [BibTex]


no image
Krylov Subspace Recycling for Fast Iterative Least-Squares in Machine Learning

Roos, F. D., Hennig, P.

arXiv preprint arXiv:1706.00241, 2017 (article)

Abstract
Solving symmetric positive definite linear problems is a fundamental computational task in machine learning. The exact solution, famously, is cubicly expensive in the size of the matrix. To alleviate this problem, several linear-time approximations, such as spectral and inducing-point methods, have been suggested and are now in wide use. These are low-rank approximations that choose the low-rank space a priori and do not refine it over time. While this allows linear cost in the data-set size, it also causes a finite, uncorrected approximation error. Authors from numerical linear algebra have explored ways to iteratively refine such low-rank approximations, at a cost of a small number of matrix-vector multiplications. This idea is particularly interesting in the many situations in machine learning where one has to solve a sequence of related symmetric positive definite linear problems. From the machine learning perspective, such deflation methods can be interpreted as transfer learning of a low-rank approximation across a time-series of numerical tasks. We study the use of such methods for our field. Our empirical results show that, on regression and classification problems of intermediate size, this approach can interpolate between low computational cost and numerical precision.

pn

link (url) Project Page [BibTex]


no image
Chemically active colloids near osmotic-responsive walls with surface-chemistry gradients

Popescu, M. N., Uspal, W. E., Dietrich, S.

Journal of Physics: Condensed Matter, 29, IOP Publishing, Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Active colloids in the context of chemical kinetics

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

Journal of Physics A, 50, IOP Pub., Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Microbots Decorated with Silver Nanoparticles Kill Bacteria in Aqueous Media

Vilela, D., Stanton, M. M., Parmar, J., Sánchez, S.

ACS Applied Materials and Interfaces, 9(27):22093-22100, American Chemical Society, Washington, DC, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Measurement of second-order response without perturbation

Helden, L., Basu, U., Krüger, M., Bechinger, C.

EPL, 116(6), IOP Publishing, Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Swimming with a cage: low-Reynolds-number locomotion inside a droplet

Reigh, S., Zhu, L. L., Gallaire, F., Lauga, E.

Soft Matter, 13(17):3161-3173, Royal Society of Chemistry, Cambridge, UK, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Active Particle Accumulation at Boundaries: A Strategy to Measure Contact Angles

Simmchen, J., Malgaretti, P.

ChemNanoMat, 3(11):790-793, Wiley, Weinheim, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Two-fluid model for locomotion under self-confinement

Reigh, S., Lauga, E.

Physical Review Fluids, 2(9), American Physical Society, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Bubble gating in biological ion channels: A density functional theory study

Gu\ssmann, Florian, Roth, R.

Physical Review E, 95(6), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Nematic films at chemically structured surfaces

Silvestre, N. M., Telo da Gama, M. M., Tasinkevych, M.

Journal of Physics: Condensed Matter, 29(7), IOP Publishing, Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Nonadditive interactions and phase transitions in strongly confined colloidal systems

Vasilyev, O., Dietrich, S., Kondrat, S.

Soft Matter, 14(4):586-596, Royal Society of Chemistry, Cambridge, UK, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Platinum-free cobalt ferrite based micromotors for antibiotic removal

Parmar, J., Villa, K., Vilela, D., Sánchez, S.

Applied Materials Today, 9, pages: 605-611, Elsevier, Amsterdam, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Stationary and time-dependent heat transfer in paradigmatic many-body geometries

Asheichyk, Kiryl

Universität Stuttgart, Stuttgart, 2017 (mastersthesis)

icm

[BibTex]

[BibTex]


no image
Perils of ad hoc approximations for the activity function of chemically powered colloids

Popescu, M. N., Uspal, W. E., Tasinkevych, M., Dietrich, S.

The European Physical Journal E, 40, EDP Sciences; Società Italiana di Fisica; Springer, Les Ulis; Bologna; Heidelberg, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Transient Casimir Forces from Quenches in Thermal and Active Matter

Rohwer, C. M., Kardar, M., Krüger, M.

Physical Review Letters, 118(1), American Physical Society, Woodbury, N.Y., 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Rotational motion of dimers of Janus particles

Majee, A.

European Physical Journal E, 40(3), Springer, Berlin, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Designing Micro- and Nanoswimmers for Specific Applications

Katuri, J., Ma, X., Stanton, M. M., Sanchez, S.

Accounts of Chemical Research, 50(1):2-11, American Chemical Society, Easton, Pa., 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Wedge wetting by electrolyte solutions

Mußotter, M., Bier, M.

Physical Review E, 96(3), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Effect of boundaries on vacuum field fluctuations and radiation-mediated interactions between atoms

Armata, F., Butera, S., Fiscelli, G., Incardone, R., Notararigo, V., Palacino, R., Passante, R., Rizzuto, L., Spagnolo, S.

Journal of Physics: Conference Series, 880, IOP Publishing, Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Non-equilibrium forces after temperature quenches in ideal fluids with conserved density

Hölzl, Christian

Universität Stuttgart, Stuttgart, 2017 (mastersthesis)

icm

[BibTex]

[BibTex]


no image
Numerical studies of active colloids at fluid interfaces

Peter, Toni

Universität Stuttgart, Stuttgart, 2017 (mastersthesis)

icm

[BibTex]

[BibTex]


no image
Smectic phases in ionic liquid crystals

Bartsch, H., Bier, M., Dietrich, S.

Journal of Physics: Condensed Matter, 29(46), IOP Publishing, Bristol, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Pushing Bacterial Biohybrids to In Vivo Applications

Stanton, M. M., Sánchez, S.

Trends in Biotechnology, 35(10):910-913, Elsevier Current Trends, Amsterdam, Netherlands, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
New Directions for Learning with Kernels and Gaussian Processes (Dagstuhl Seminar 16481)

Gretton, A., Hennig, P., Rasmussen, C., Schölkopf, B.

Dagstuhl Reports, 6(11):142-167, 2017 (book)

ei pn

DOI [BibTex]

DOI [BibTex]


no image
Particles with nonlinear electric response: Suppressing van der Waals forces by an external field

Soo, H., Dean, D. S., Krüger, M.

Physical Review E, 95(1), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Electrolyte solutions at curved electrodes. I. Mesoscopic approach

Reindl, A., Bier, M., Dietrich, S.

The Journal of Chemical Physics, 146(15), American Institute of Physics, Woodbury, N.Y., 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Self-propelling micro-nanorobots: challenges and fututre perspectives in nanomedicine

Ma, X., Sánchez, S.

Nanomedicine, 12(12):1363-1367, Future Medicine Ltd, London, UK, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Salt-induced microheterogeneities in binary liquid mixtures

Bier, M., Mars, J., Li, H., Mezger, M.

Physical Review E, 96(2), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Forward-backward multiplicity correlations in proton-proton collisions from several GeV to LHC energies

Bravina, L., Bleibel, J., Emaus, R., Zabrodin, E.

EPJ Web of Conferences, 164, EDP Sciences, Les Ulis, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Key-lock colloids in a nematic liquid crystal

Silvestre, N. M., Tasinkevych, M.

Physical Review E, 95(1), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Collective dynamics of laterally confined active particles near fluid-fluid interfaces

Kistner, Irina

Universität Stuttgart, Stuttgart, 2017 (mastersthesis)

icm

[BibTex]

[BibTex]


no image
Fluctuation induced forces in critical films with disorder at their surfaces

Maciolek, A., Vasilyev, O., Dotsenko, V., Dietrich, S.

Journal of Statistical Mechanics: Theory and Experiment, Institute of Physics Publishing, Bristol, England, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Action at a distance in classical uniaxial ferromagnetic arrays

Abraham, D. B., Maciolek, A., Squarcini, A., Vasilyev, O.

Physical Review E, 96(4), American Physical Society, Melville, NY, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Cooperative behavior of biased probes in crowded interacting systems

Vasilyev, O., Bénichou, O., Mej\’\ia-Monasterio, C., Weeks, E. R., Oshanin, G.

Soft Matter, 13(41):7617-7624, Royal Society of Chemistry, Cambridge, UK, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Nonequilibrium Fluctuational Quantum Electrodynamics: Heat Radiation, Heat Transfer, and Force

Bimonte, G., Emig, T., Kardar, M., Krüger, M.

Annual Review of Condensed Matter Physics, 8, pages: 119-143, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
A Gaussian theory for fluctuations in simple liquids

Krüger, M., Dean, D. S.

The Journal of Chemical Physics, 146(13), American Institute of Physics, Woodbury, N.Y., 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Critical Casimir interactions between colloids around the critical point of binary solvents

Stuij, S. G., Labbe-Laurent, M., Kodger, T. E., Maciolek, A., Schall, P.

Soft Matter, 13(31):5233-5249, Royal Society of Chemistry, Cambridge, UK, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]


no image
Bio-catalytic mesoporous Janus nano-motors powered by catalase enzyme

Ma, X., Sánchez, S.

Tetrahedron, 73(33):4883-4886, Elsevier Science, Kidlington, 2017 (article)

icm

DOI [BibTex]

DOI [BibTex]