Header logo is


2018


no image
Parallel and functionally segregated processing of task phase and conscious content in the prefrontal cortex

Kapoor, V., Besserve, M., Logothetis, N. K., Panagiotaropoulos, T. I.

Communications Biology, 1(215):1-12, December 2018 (article)

ei

link (url) DOI [BibTex]

2018


link (url) DOI [BibTex]


Thumb xl screen shot 2019 01 07 at 12.05.00
Control of Musculoskeletal Systems using Learned Dynamics Models

Büchler, D., Calandra, R., Schölkopf, B., Peters, J.

IEEE Robotics and Automation Letters, Robotics and Automation Letters, 3(4):3161-3168, IEEE, 2018 (article)

Abstract
Controlling musculoskeletal systems, especially robots actuated by pneumatic artificial muscles, is a challenging task due to nonlinearities, hysteresis effects, massive actuator de- lay and unobservable dependencies such as temperature. Despite such difficulties, muscular systems offer many beneficial prop- erties to achieve human-comparable performance in uncertain and fast-changing tasks. For example, muscles are backdrivable and provide variable stiffness while offering high forces to reach high accelerations. In addition, the embodied intelligence deriving from the compliance might reduce the control demands for specific tasks. In this paper, we address the problem of how to accurately control musculoskeletal robots. To address this issue, we propose to learn probabilistic forward dynamics models using Gaussian processes and, subsequently, to employ these models for control. However, Gaussian processes dynamics models cannot be set-up for our musculoskeletal robot as for traditional motor- driven robots because of unclear state composition etc. We hence empirically study and discuss in detail how to tune these approaches to complex musculoskeletal robots and their specific challenges. Moreover, we show that our model can be used to accurately control an antagonistic pair of pneumatic artificial muscles for a trajectory tracking task while considering only one- step-ahead predictions of the forward model and incorporating model uncertainty.

ei

RAL18final link (url) DOI [BibTex]

RAL18final link (url) DOI [BibTex]


no image
Infinite Factorial Finite State Machine for Blind Multiuser Channel Estimation

Ruiz, F. J. R., Valera, I., Svensson, L., Perez-Cruz, F.

IEEE Transactions on Cognitive Communications and Networking, 4(2):177-191, June 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Assisting Movement Training and Execution With Visual and Haptic Feedback

Ewerton, M., Rother, D., Weimar, J., Kollegger, G., Wiemeyer, J., Peters, J., Maeda, G.

Frontiers in Neurorobotics, 12, May 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


Thumb xl screenshot 2018 05 18 16 38 40
Learning 3D Shape Completion under Weak Supervision

Stutz, D., Geiger, A.

Arxiv, May 2018 (article)

Abstract
We address the problem of 3D shape completion from sparse and noisy point clouds, a fundamental problem in computer vision and robotics. Recent approaches are either data-driven or learning-based: Data-driven approaches rely on a shape model whose parameters are optimized to fit the observations; Learning-based approaches, in contrast, avoid the expensive optimization step by learning to directly predict complete shapes from incomplete observations in a fully-supervised setting. However, full supervision is often not available in practice. In this work, we propose a weakly-supervised learning-based approach to 3D shape completion which neither requires slow optimization nor direct supervision. While we also learn a shape prior on synthetic data, we amortize, i.e., learn, maximum likelihood fitting using deep neural networks resulting in efficient shape completion without sacrificing accuracy. On synthetic benchmarks based on ShapeNet and ModelNet as well as on real robotics data from KITTI and Kinect, we demonstrate that the proposed amortized maximum likelihood approach is able to compete with fully supervised baselines and outperforms data-driven approaches, while requiring less supervision and being significantly faster.

avg

PDF Project Page Project Page [BibTex]


no image
Mixture of Attractors: A Novel Movement Primitive Representation for Learning Motor Skills From Demonstrations

Manschitz, S., Gienger, M., Kober, J., Peters, J.

IEEE Robotics and Automation Letters, 3(2):926-933, April 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Probabilistic movement primitives under unknown system dynamics

Paraschos, A., Rueckert, E., Peters, J., Neumann, G.

Advanced Robotics, 32(6):297-310, April 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
An Algorithmic Perspective on Imitation Learning

Osa, T., Pajarinen, J., Neumann, G., Bagnell, J., Abbeel, P., Peters, J.

Foundations and Trends in Robotics, 7(1-2):1-179, March 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Using Probabilistic Movement Primitives in Robotics

Paraschos, A., Daniel, C., Peters, J., Neumann, G.

Autonomous Robots, 42(3):529-551, March 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
A kernel-based approach to learning contact distributions for robot manipulation tasks

Kroemer, O., Leischnig, S., Luettgen, S., Peters, J.

Autonomous Robots, 42(3):581-600, March 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Approximate Value Iteration Based on Numerical Quadrature

Vinogradska, J., Bischoff, B., Peters, J.

IEEE Robotics and Automation Letters, 3(2):1330-1337, January 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Biomimetic Tactile Sensors and Signal Processing with Spike Trains: A Review

Yi, Z., Zhang, Y., Peters, J.

Sensors and Actuators A: Physical, 269, pages: 41-52, January 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Design and Analysis of the NIPS 2016 Review Process

Shah*, N., Tabibian*, B., Muandet, K., Guyon, I., von Luxburg, U.

Journal of Machine Learning Research, 19(49):1-34, 2018, *equal contribution (article)

ei slt

arXiv link (url) [BibTex]

arXiv link (url) [BibTex]


no image
Intrinsic Motivation and Mental Replay enable Efficient Online Adaptation in Stochastic Recurrent Networks

Tanneberg, D., Peters, J., Rueckert, E.

Neural Networks, 109, pages: 67-80, 2018 (article)

ei

link (url) DOI [BibTex]


no image
A Flexible Approach for Fair Classification

Zafar, M. B., Valera, I., Gomez Rodriguez, M., Gummadi, K.

Journal of Machine Learning, 2018 (article) Accepted

ei

[BibTex]

[BibTex]


no image
Adaptation and Robust Learning of Probabilistic Movement Primitives

Gomez-Gonzalez, S., Neumann, G., Schölkopf, B., Peters, J.

IEEE Transactions on Robotics, 2018 (article) In revision

ei

arXiv [BibTex]

arXiv [BibTex]


no image
Does universal controllability of physical systems prohibit thermodynamic cycles?

Janzing, D., Wocjan, P.

Open Systems and Information Dynamics, 25(3):1850016, 2018 (article)

ei

PDF DOI [BibTex]

PDF DOI [BibTex]


no image
Learning Causality and Causality-Related Learning: Some Recent Progress

Zhang, K., Schölkopf, B., Spirtes, P., Glymour, C.

National Science Review, 5(1):26-29, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


Thumb xl hassan teaser paper
Augmented Reality Meets Computer Vision: Efficient Data Generation for Urban Driving Scenes

Alhaija, H., Mustikovela, S., Mescheder, L., Geiger, A., Rother, C.

International Journal of Computer Vision (IJCV), 2018, 2018 (article)

Abstract
The success of deep learning in computer vision is based on the availability of large annotated datasets. To lower the need for hand labeled images, virtually rendered 3D worlds have recently gained popularity. Unfortunately, creating realistic 3D content is challenging on its own and requires significant human effort. In this work, we propose an alternative paradigm which combines real and synthetic data for learning semantic instance segmentation and object detection models. Exploiting the fact that not all aspects of the scene are equally important for this task, we propose to augment real-world imagery with virtual objects of the target category. Capturing real-world images at large scale is easy and cheap, and directly provides real background appearances without the need for creating complex 3D models of the environment. We present an efficient procedure to augment these images with virtual objects. In contrast to modeling complete 3D environments, our data augmentation approach requires only a few user interactions in combination with 3D models of the target object category. Leveraging our approach, we introduce a novel dataset of augmented urban driving scenes with 360 degree images that are used as environment maps to create realistic lighting and reflections on rendered objects. We analyze the significance of realistic object placement by comparing manual placement by humans to automatic methods based on semantic scene analysis. This allows us to create composite images which exhibit both realistic background appearance as well as a large number of complex object arrangements. Through an extensive set of experiments, we conclude the right set of parameters to produce augmented data which can maximally enhance the performance of instance segmentation models. Further, we demonstrate the utility of the proposed approach on training standard deep models for semantic instance segmentation and object detection of cars in outdoor driving scenarios. We test the models trained on our augmented data on the KITTI 2015 dataset, which we have annotated with pixel-accurate ground truth, and on the Cityscapes dataset. Our experiments demonstrate that the models trained on augmented imagery generalize better than those trained on fully synthetic data or models trained on limited amounts of annotated real data.

avg

pdf Project Page [BibTex]

pdf Project Page [BibTex]


no image
Online optimal trajectory generation for robot table tennis

Koc, O., Maeda, G., Peters, J.

Robotics and Autonomous Systems, 105, pages: 121-137, 2018 (article)

ei

PDF link (url) DOI [BibTex]

PDF link (url) DOI [BibTex]


no image
Counterfactual Mean Embedding: A Kernel Method for Nonparametric Causal Inference

Muandet, K., Kanagawa, M., Saengkyongam, S., Marukata, S.

Arxiv e-prints, arXiv:1805.08845v1 [stat.ML], 2018 (article)

Abstract
This paper introduces a novel Hilbert space representation of a counterfactual distribution---called counterfactual mean embedding (CME)---with applications in nonparametric causal inference. Counterfactual prediction has become an ubiquitous tool in machine learning applications, such as online advertisement, recommendation systems, and medical diagnosis, whose performance relies on certain interventions. To infer the outcomes of such interventions, we propose to embed the associated counterfactual distribution into a reproducing kernel Hilbert space (RKHS) endowed with a positive definite kernel. Under appropriate assumptions, the CME allows us to perform causal inference over the entire landscape of the counterfactual distribution. The CME can be estimated consistently from observational data without requiring any parametric assumption about the underlying distributions. We also derive a rate of convergence which depends on the smoothness of the conditional mean and the Radon-Nikodym derivative of the underlying marginal distributions. Our framework can deal with not only real-valued outcome, but potentially also more complex and structured outcomes such as images, sequences, and graphs. Lastly, our experimental results on off-policy evaluation tasks demonstrate the advantages of the proposed estimator.

ei pn

arXiv [BibTex]

arXiv [BibTex]


no image
Hierarchical Reinforcement Learning of Multiple Grasping Strategies with Human Instructions

Osa, T., Peters, J., Neumann, G.

Advanced Robotics, 32(18):955-968, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Case series: Slowing alpha rhythm in late-stage ALS patients

Hohmann, M. R., Fomina, T., Jayaram, V., Emde, T., Just, J., Synofzik, M., Schölkopf, B., Schöls, L., Grosse-Wentrup, M.

Clinical Neurophysiology, 129(2):406-408, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Inverse Reinforcement Learning via Nonparametric Spatio-Temporal Subgoal Modeling

Šošić, A., Rueckert, E., Peters, J., Zoubir, A., Koeppl, H.

Journal of Machine Learning Research, 19(69):1-45, 2018 (article)

ei

link (url) [BibTex]

link (url) [BibTex]


no image
Grip Stabilization of Novel Objects using Slip Prediction

Veiga, F., Peters, J., Hermans, T.

IEEE Transactions on Haptics, 2018 (article) In press

ei

DOI [BibTex]

DOI [BibTex]


no image
Electrophysiological correlates of neurodegeneration in motor and non-motor brain regions in amyotrophic lateral sclerosis—implications for brain–computer interfacing

Kellmeyer, P., Grosse-Wentrup, M., Schulze-Bonhage, A., Ziemann, U., Ball, T.

Journal of Neural Engineering, 15(4):041003, IOP Publishing, 2018 (article)

ei

link (url) [BibTex]

link (url) [BibTex]


no image
Autofocusing-based phase correction

Loktyushin, A., Ehses, P., Schölkopf, B., Scheffler, K.

Magnetic Resonance in Medicine, 80(3):958-968, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Prediction of Glucose Tolerance without an Oral Glucose Tolerance Test

Babbar, R., Heni, M., Peter, A., Hrabě de Angelis, M., Häring, H., Fritsche, A., Preissl, H., Schölkopf, B., Wagner, R.

Frontiers in Endocrinology, 9, pages: 82, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Invariant Models for Causal Transfer Learning

Rojas-Carulla, M., Schölkopf, B., Turner, R., Peters, J.

Journal of Machine Learning Research, 19(36):1-34, 2018 (article)

ei

link (url) [BibTex]

link (url) [BibTex]


no image
MOABB: Trustworthy algorithm benchmarking for BCIs

Jayaram, V., Barachant, A.

Journal of Neural Engineering, 15(6):066011, 2018 (article)

ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
f-Divergence constrained policy improvement

Belousov, B., Peters, J.

Journal of Machine Learning Research, 2018 (article) Submitted

ei

[BibTex]

[BibTex]


no image
Phylogenetic convolutional neural networks in metagenomics

Fioravanti*, D., Giarratano*, Y., Maggio*, V., Agostinelli, C., Chierici, M., Jurman, G., Furlanello, C.

BMC Bioinformatics, 19(2):49 pages, 2018, *equal contribution (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Food specific inhibitory control under negative mood in binge-eating disorder: Evidence from a multimethod approach

Leehr, E. J., Schag, K., Dresler, T., Grosse-Wentrup, M., Hautzinger, M., Fallgatter, A. J., Zipfel, S., Giel, K. E., Ehlis, A.

International Journal of Eating Disorders, 51(2):112-123, Wiley Online Library, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Linking imaging to omics utilizing image-guided tissue extraction

Disselhorst, J. A., Krueger, M. A., Ud-Dean, S. M. M., Bezrukov, I., Jarboui, M. A., Trautwein, C., Traube, A., Spindler, C., Cotton, J. M., Leibfritz, D., Pichler, B. J.

Proceedings of the National Academy of Sciences, 115(13):E2980-E2987, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Kernel-based tests for joint independence

Pfister, N., Bühlmann, P., Schölkopf, B., Peters, J.

Journal of the Royal Statistical Society: Series B (Statistical Methodology), 80(1):5-31, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


Thumb xl stutz
Learning 3D Shape Completion under Weak Supervision

Stutz, D., Geiger, A.

International Journal of Computer Vision (IJCV), 2018, 2018 (article)

Abstract
We address the problem of 3D shape completion from sparse and noisy point clouds, a fundamental problem in computer vision and robotics. Recent approaches are either data-driven or learning-based: Data-driven approaches rely on a shape model whose parameters are optimized to fit the observations; Learning-based approaches, in contrast, avoid the expensive optimization step by learning to directly predict complete shapes from incomplete observations in a fully-supervised setting. However, full supervision is often not available in practice. In this work, we propose a weakly-supervised learning-based approach to 3D shape completion which neither requires slow optimization nor direct supervision. While we also learn a shape prior on synthetic data, we amortize, i.e., learn, maximum likelihood fitting using deep neural networks resulting in efficient shape completion without sacrificing accuracy. On synthetic benchmarks based on ShapeNet and ModelNet as well as on real robotics data from KITTI and Kinect, we demonstrate that the proposed amortized maximum likelihood approach is able to compete with a fully supervised baseline and outperforms the data-driven approach of Engelmann et al., while requiring less supervision and being significantly faster.

avg

pdf Project Page [BibTex]

pdf Project Page [BibTex]


no image
Discriminative Transfer Learning for General Image Restoration

Xiao, L., Heide, F., Heidrich, W., Schölkopf, B., Hirsch, M.

IEEE Transactions on Image Processing, 27(8):4091-4104, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Dissecting the synapse- and frequency-dependent network mechanisms of in vivo hippocampal sharp wave-ripples

Ramirez-Villegas, J. F., Willeke, K. F., Logothetis, N. K., Besserve, M.

Neuron, 100(5):1224-1240, 2018 (article)

ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Optimizing Execution of Dynamic Goal-Directed Robot Movements with Learning Control

Koc, O., Maeda, G., Peters, J.

IEEE Transactions on Robotics, 2018 (article) Submitted

ei

arXiv [BibTex]

arXiv [BibTex]


no image
Visualizing and understanding Sum-Product Networks

Vergari, A., Di Mauro, N., Esposito, F.

Machine Learning, 2018 (article)

ei

DOI [BibTex]

DOI [BibTex]


no image
Learning to serve: an experimental study for a new learning from demonstrations framework

Koc, O., Peters, J.

IEEE Robotics and Automation Letters (ICRA/RA-L), 2018 (article) Accepted

ei

[BibTex]

[BibTex]


Thumb xl objectflow
Object Scene Flow

Menze, M., Heipke, C., Geiger, A.

ISPRS Journal of Photogrammetry and Remote Sensing, 2018 (article)

Abstract
This work investigates the estimation of dense three-dimensional motion fields, commonly referred to as scene flow. While great progress has been made in recent years, large displacements and adverse imaging conditions as observed in natural outdoor environments are still very challenging for current approaches to reconstruction and motion estimation. In this paper, we propose a unified random field model which reasons jointly about 3D scene flow as well as the location, shape and motion of vehicles in the observed scene. We formulate the problem as the task of decomposing the scene into a small number of rigidly moving objects sharing the same motion parameters. Thus, our formulation effectively introduces long-range spatial dependencies which commonly employed local rigidity priors are lacking. Our inference algorithm then estimates the association of image segments and object hypotheses together with their three-dimensional shape and motion. We demonstrate the potential of the proposed approach by introducing a novel challenging scene flow benchmark which allows for a thorough comparison of the proposed scene flow approach with respect to various baseline models. In contrast to previous benchmarks, our evaluation is the first to provide stereo and optical flow ground truth for dynamic real-world urban scenes at large scale. Our experiments reveal that rigid motion segmentation can be utilized as an effective regularizer for the scene flow problem, improving upon existing two-frame scene flow methods. At the same time, our method yields plausible object segmentations without requiring an explicitly trained recognition model for a specific object class.

avg

Project Page [BibTex]

Project Page [BibTex]


no image
In-Hand Object Stabilization by Independent Finger Control

Veiga, F. F., Edin, B. B., Peters, J.

IEEE Transactions on Robotics, 2018 (article) Submitted

ei

[BibTex]

[BibTex]


no image
Non-Equilibrium Relations for Bounded Rational Decision-Making in Changing Environments

Grau-Moya, J, Krüger, M, Braun, DA

Entropy, 20(1:1):1-28, January 2018 (article)

Abstract
Living organisms from single cells to humans need to adapt continuously to respond to changes in their environment. The process of behavioural adaptation can be thought of as improving decision-making performance according to some utility function. Here, we consider an abstract model of organisms as decision-makers with limited information-processing resources that trade off between maximization of utility and computational costs measured by a relative entropy, in a similar fashion to thermodynamic systems undergoing isothermal transformations. Such systems minimize the free energy to reach equilibrium states that balance internal energy and entropic cost. When there is a fast change in the environment, these systems evolve in a non-equilibrium fashion because they are unable to follow the path of equilibrium distributions. Here, we apply concepts from non-equilibrium thermodynamics to characterize decision-makers that adapt to changing environments under the assumption that the temporal evolution of the utility function is externally driven and does not depend on the decision-maker’s action. This allows one to quantify performance loss due to imperfect adaptation in a general manner and, additionally, to find relations for decision-making similar to Crooks’ fluctuation theorem and Jarzynski’s equality. We provide simulations of several exemplary decision and inference problems in the discrete and continuous domains to illustrate the new relations.

ei

DOI [BibTex]

DOI [BibTex]

2011


no image
Causal Inference on Discrete Data using Additive Noise Models

Peters, J., Janzing, D., Schölkopf, B.

IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(12):2436-2450, December 2011 (article)

Abstract
Inferring the causal structure of a set of random variables from a finite sample of the joint distribution is an important problem in science. The case of two random variables is particularly challenging since no (conditional) independences can be exploited. Recent methods that are based on additive noise models suggest the following principle: Whenever the joint distribution {\bf P}^{(X,Y)} admits such a model in one direction, e.g., Y=f(X)+N, N \perp\kern-6pt \perp X, but does not admit the reversed model X=g(Y)+\tilde{N}, \tilde{N} \perp\kern-6pt \perp Y, one infers the former direction to be causal (i.e., X\rightarrow Y). Up to now, these approaches only dealt with continuous variables. In many situations, however, the variables of interest are discrete or even have only finitely many states. In this work, we extend the notion of additive noise models to these cases. We prove that it almost never occurs that additive noise models can be fit in both directions. We further propose an efficient algorithm that is able to perform this way of causal inference on finite samples of discrete variables. We show that the algorithm works on both synthetic and real data sets.

ei

PDF Web DOI [BibTex]

2011


PDF Web DOI [BibTex]


no image
Spontaneous epigenetic variation in the Arabidopsis thaliana methylome

Becker, C., Hagmann, J., Müller, J., Koenig, D., Stegle, O., Borgwardt, K., Weigel, D.

Nature, 480(7376):245-249, December 2011 (article)

Abstract
Heritable epigenetic polymorphisms, such as differential cytosine methylation, can underlie phenotypic variation1, 2. Moreover, wild strains of the plant Arabidopsis thaliana differ in many epialleles3, 4, and these can influence the expression of nearby genes1, 2. However, to understand their role in evolution5, it is imperative to ascertain the emergence rate and stability of epialleles, including those that are not due to structural variation. We have compared genome-wide DNA methylation among 10 A. thaliana lines, derived 30 generations ago from a common ancestor6. Epimutations at individual positions were easily detected, and close to 30,000 cytosines in each strain were differentially methylated. In contrast, larger regions of contiguous methylation were much more stable, and the frequency of changes was in the same low range as that of DNA mutations7. Like individual positions, the same regions were often affected by differential methylation in independent lines, with evidence for recurrent cycles of forward and reverse mutations. Transposable elements and short interfering RNAs have been causally linked to DNA methylation8. In agreement, differentially methylated sites were farther from transposable elements and showed less association with short interfering RNA expression than invariant positions. The biased distribution and frequent reversion of epimutations have important implications for the potential contribution of sequence-independent epialleles to plant evolution.

ei

Web DOI [BibTex]

Web DOI [BibTex]


no image
HHfrag: HMM-based fragment detection using HHpred

Kalev, I., Habeck, M.

Bioinformatics, 27(22):3110-3116, November 2011 (article)

Abstract
Motivation: Over the last decade, both static and dynamic fragment libraries for protein structure prediction have been introduced. The former are built from clusters in either sequence or structure space and aim to extract a universal structural alphabet. The latter are tailored for a particular query protein sequence and aim to provide local structural templates that need to be assembled in order to build the full-length structure. Results: Here, we introduce HHfrag, a dynamic HMM-based fragment search method built on the profile–profile comparison tool HHpred. We show that HHfrag provides advantages over existing fragment assignment methods in that it: (i) improves the precision of the fragments at the expense of a minor loss in sequence coverage; (ii) detects fragments of variable length (6–21 amino acid residues); (iii) allows for gapped fragments and (iv) does not assign fragments to regions where there is no clear sequence conservation. We illustrate the usefulness of fragments detected by HHfrag on targets from most recent CASP.

ei

Web DOI [BibTex]

Web DOI [BibTex]


no image
Reward-Weighted Regression with Sample Reuse for Direct Policy Search in Reinforcement Learning

Hachiya, H., Peters, J., Sugiyama, M.

Neural Computation, 23(11):2798-2832, November 2011 (article)

Abstract
Direct policy search is a promising reinforcement learning framework, in particular for controlling continuous, high-dimensional systems. Policy search often requires a large number of samples for obtaining a stable policy update estimator, and this is prohibitive when the sampling cost is expensive. In this letter, we extend an expectation-maximization-based policy search method so that previously collected samples can be efficiently reused. The usefulness of the proposed method, reward-weighted regression with sample reuse (R), is demonstrated through robot learning experiments.

ei

Web DOI [BibTex]