Header logo is


2018


Deep Reinforcement Learning for Event-Triggered Control
Deep Reinforcement Learning for Event-Triggered Control

Baumann, D., Zhu, J., Martius, G., Trimpe, S.

In Proceedings of the 57th IEEE International Conference on Decision and Control (CDC), pages: 943-950, 57th IEEE International Conference on Decision and Control (CDC), December 2018 (inproceedings)

al ics

arXiv PDF DOI Project Page Project Page [BibTex]

2018


arXiv PDF DOI Project Page Project Page [BibTex]


Customized Multi-Person Tracker
Customized Multi-Person Tracker

Ma, L., Tang, S., Black, M. J., Van Gool, L.

In Computer Vision – ACCV 2018, Springer International Publishing, Asian Conference on Computer Vision, December 2018 (inproceedings)

ps

PDF Project Page [BibTex]

PDF Project Page [BibTex]


On the Integration of Optical Flow and Action Recognition
On the Integration of Optical Flow and Action Recognition

Sevilla-Lara, L., Liao, Y., Güney, F., Jampani, V., Geiger, A., Black, M. J.

In German Conference on Pattern Recognition (GCPR), LNCS 11269, pages: 281-297, Springer, Cham, October 2018 (inproceedings)

Abstract
Most of the top performing action recognition methods use optical flow as a "black box" input. Here we take a deeper look at the combination of flow and action recognition, and investigate why optical flow is helpful, what makes a flow method good for action recognition, and how we can make it better. In particular, we investigate the impact of different flow algorithms and input transformations to better understand how these affect a state-of-the-art action recognition method. Furthermore, we fine tune two neural-network flow methods end-to-end on the most widely used action recognition dataset (UCF101). Based on these experiments, we make the following five observations: 1) optical flow is useful for action recognition because it is invariant to appearance, 2) optical flow methods are optimized to minimize end-point-error (EPE), but the EPE of current methods is not well correlated with action recognition performance, 3) for the flow methods tested, accuracy at boundaries and at small displacements is most correlated with action recognition performance, 4) training optical flow to minimize classification error instead of minimizing EPE improves recognition performance, and 5) optical flow learned for the task of action recognition differs from traditional optical flow especially inside the human body and at the boundary of the body. These observations may encourage optical flow researchers to look beyond EPE as a goal and guide action recognition researchers to seek better motion cues, leading to a tighter integration of the optical flow and action recognition communities.

avg ps

arXiv DOI [BibTex]

arXiv DOI [BibTex]


Temporal Interpolation as an Unsupervised Pretraining Task for Optical Flow Estimation
Temporal Interpolation as an Unsupervised Pretraining Task for Optical Flow Estimation

Wulff, J., Black, M. J.

In German Conference on Pattern Recognition (GCPR), LNCS 11269, pages: 567-582, Springer, Cham, October 2018 (inproceedings)

Abstract
The difficulty of annotating training data is a major obstacle to using CNNs for low-level tasks in video. Synthetic data often does not generalize to real videos, while unsupervised methods require heuristic n losses. Proxy tasks can overcome these issues, and start by training a network for a task for which annotation is easier or which can be trained unsupervised. The trained network is then fine-tuned for the original task using small amounts of ground truth data. Here, we investigate frame interpolation as a proxy task for optical flow. Using real movies, we train a CNN unsupervised for temporal interpolation. Such a network implicitly estimates motion, but cannot handle untextured regions. By fi ne-tuning on small amounts of ground truth flow, the network can learn to fill in homogeneous regions and compute full optical flow fi elds. Using this unsupervised pre-training, our network outperforms similar architectures that were trained supervised using synthetic optical flow.

ps

pdf arXiv DOI Project Page [BibTex]

pdf arXiv DOI Project Page [BibTex]


Human Motion Parsing by Hierarchical Dynamic Clustering
Human Motion Parsing by Hierarchical Dynamic Clustering

Zhang, Y., Tang, S., Sun, H., Neumann, H.

In Proceedings of the British Machine Vision Conference (BMVC), pages: 269, BMVA Press, 29th British Machine Vision Conference, September 2018 (inproceedings)

Abstract
Parsing continuous human motion into meaningful segments plays an essential role in various applications. In this work, we propose a hierarchical dynamic clustering framework to derive action clusters from a sequence of local features in an unsuper- vised bottom-up manner. We systematically investigate the modules in this framework and particularly propose diverse temporal pooling schemes, in order to realize accurate temporal action localization. We demonstrate our method on two motion parsing tasks: temporal action segmentation and abnormal behavior detection. The experimental results indicate that the proposed framework is significantly more effective than the other related state-of-the-art methods on several datasets.

ps

pdf Project Page [BibTex]

pdf Project Page [BibTex]


Generating {3D} Faces using Convolutional Mesh Autoencoders
Generating 3D Faces using Convolutional Mesh Autoencoders

Ranjan, A., Bolkart, T., Sanyal, S., Black, M. J.

In European Conference on Computer Vision (ECCV), Lecture Notes in Computer Science, vol 11207, pages: 725-741, Springer, Cham, September 2018 (inproceedings)

Abstract
Learned 3D representations of human faces are useful for computer vision problems such as 3D face tracking and reconstruction from images, as well as graphics applications such as character generation and animation. Traditional models learn a latent representation of a face using linear subspaces or higher-order tensor generalizations. Due to this linearity, they can not capture extreme deformations and non-linear expressions. To address this, we introduce a versatile model that learns a non-linear representation of a face using spectral convolutions on a mesh surface. We introduce mesh sampling operations that enable a hierarchical mesh representation that captures non-linear variations in shape and expression at multiple scales within the model. In a variational setting, our model samples diverse realistic 3D faces from a multivariate Gaussian distribution. Our training data consists of 20,466 meshes of extreme expressions captured over 12 different subjects. Despite limited training data, our trained model outperforms state-of-the-art face models with 50% lower reconstruction error, while using 75% fewer parameters. We also show that, replacing the expression space of an existing state-of-the-art face model with our autoencoder, achieves a lower reconstruction error. Our data, model and code are available at http://coma.is.tue.mpg.de/.

ps

Code (tensorflow) Code (pytorch) Project Page paper supplementary DOI Project Page Project Page [BibTex]

Code (tensorflow) Code (pytorch) Project Page paper supplementary DOI Project Page Project Page [BibTex]


Part-Aligned Bilinear Representations for Person Re-identification
Part-Aligned Bilinear Representations for Person Re-identification

Suh, Y., Wang, J., Tang, S., Mei, T., Lee, K. M.

In European Conference on Computer Vision (ECCV), 11218, pages: 418-437, Springer, Cham, September 2018 (inproceedings)

Abstract
Comparing the appearance of corresponding body parts is essential for person re-identification. However, body parts are frequently misaligned be- tween detected boxes, due to the detection errors and the pose/viewpoint changes. In this paper, we propose a network that learns a part-aligned representation for person re-identification. Our model consists of a two-stream network, which gen- erates appearance and body part feature maps respectively, and a bilinear-pooling layer that fuses two feature maps to an image descriptor. We show that it results in a compact descriptor, where the inner product between two image descriptors is equivalent to an aggregation of the local appearance similarities of the cor- responding body parts, and thereby significantly reduces the part misalignment problem. Our approach is advantageous over other pose-guided representations by learning part descriptors optimal for person re-identification. Training the net- work does not require any part annotation on the person re-identification dataset. Instead, we simply initialize the part sub-stream using a pre-trained sub-network of an existing pose estimation network and train the whole network to minimize the re-identification loss. We validate the effectiveness of our approach by demon- strating its superiority over the state-of-the-art methods on the standard bench- mark datasets including Market-1501, CUHK03, CUHK01 and DukeMTMC, and standard video dataset MARS.

ps

pdf supplementary DOI Project Page [BibTex]

pdf supplementary DOI Project Page [BibTex]


Learning Human Optical Flow
Learning Human Optical Flow

Ranjan, A., Romero, J., Black, M. J.

In 29th British Machine Vision Conference, September 2018 (inproceedings)

Abstract
The optical flow of humans is well known to be useful for the analysis of human action. Given this, we devise an optical flow algorithm specifically for human motion and show that it is superior to generic flow methods. Designing a method by hand is impractical, so we develop a new training database of image sequences with ground truth optical flow. For this we use a 3D model of the human body and motion capture data to synthesize realistic flow fields. We then train a convolutional neural network to estimate human flow fields from pairs of images. Since many applications in human motion analysis depend on speed, and we anticipate mobile applications, we base our method on SpyNet with several modifications. We demonstrate that our trained network is more accurate than a wide range of top methods on held-out test data and that it generalizes well to real image sequences. When combined with a person detector/tracker, the approach provides a full solution to the problem of 2D human flow estimation. Both the code and the dataset are available for research.

ps

video code pdf link (url) Project Page Project Page [BibTex]

video code pdf link (url) Project Page Project Page [BibTex]


Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose and Shape Estimation
Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose and Shape Estimation

(Best Student Paper Award)

Omran, M., Lassner, C., Pons-Moll, G., Gehler, P. V., Schiele, B.

In 3DV, September 2018 (inproceedings)

Abstract
Direct prediction of 3D body pose and shape remains a challenge even for highly parameterized deep learning models. Mapping from the 2D image space to the prediction space is difficult: perspective ambiguities make the loss function noisy and training data is scarce. In this paper, we propose a novel approach (Neural Body Fitting (NBF)). It integrates a statistical body model within a CNN, leveraging reliable bottom-up semantic body part segmentation and robust top-down body model constraints. NBF is fully differentiable and can be trained using 2D and 3D annotations. In detailed experiments, we analyze how the components of our model affect performance, especially the use of part segmentations as an explicit intermediate representation, and present a robust, efficiently trainable framework for 3D human pose estimation from 2D images with competitive results on standard benchmarks. Code is available at https://github.com/mohomran/neural_body_fitting

ps

arXiv code Project Page [BibTex]


Unsupervised Learning of Multi-Frame Optical Flow with Occlusions
Unsupervised Learning of Multi-Frame Optical Flow with Occlusions

Janai, J., Güney, F., Ranjan, A., Black, M. J., Geiger, A.

In European Conference on Computer Vision (ECCV), Lecture Notes in Computer Science, vol 11220, pages: 713-731, Springer, Cham, September 2018 (inproceedings)

avg ps

pdf suppmat Video Project Page DOI Project Page [BibTex]

pdf suppmat Video Project Page DOI Project Page [BibTex]


Learning an Infant Body Model from {RGB-D} Data for Accurate Full Body Motion Analysis
Learning an Infant Body Model from RGB-D Data for Accurate Full Body Motion Analysis

Hesse, N., Pujades, S., Romero, J., Black, M. J., Bodensteiner, C., Arens, M., Hofmann, U. G., Tacke, U., Hadders-Algra, M., Weinberger, R., Muller-Felber, W., Schroeder, A. S.

In Int. Conf. on Medical Image Computing and Computer Assisted Intervention (MICCAI), September 2018 (inproceedings)

Abstract
Infant motion analysis enables early detection of neurodevelopmental disorders like cerebral palsy (CP). Diagnosis, however, is challenging, requiring expert human judgement. An automated solution would be beneficial but requires the accurate capture of 3D full-body movements. To that end, we develop a non-intrusive, low-cost, lightweight acquisition system that captures the shape and motion of infants. Going beyond work on modeling adult body shape, we learn a 3D Skinned Multi-Infant Linear body model (SMIL) from noisy, low-quality, and incomplete RGB-D data. We demonstrate the capture of shape and motion with 37 infants in a clinical environment. Quantitative experiments show that SMIL faithfully represents the data and properly factorizes the shape and pose of the infants. With a case study based on general movement assessment (GMA), we demonstrate that SMIL captures enough information to allow medical assessment. SMIL provides a new tool and a step towards a fully automatic system for GMA.

ps

pdf Project page video extended arXiv version DOI Project Page [BibTex]

pdf Project page video extended arXiv version DOI Project Page [BibTex]


Deep Directional Statistics: Pose Estimation with Uncertainty Quantification
Deep Directional Statistics: Pose Estimation with Uncertainty Quantification

Prokudin, S., Gehler, P., Nowozin, S.

European Conference on Computer Vision (ECCV), September 2018 (conference)

Abstract
Modern deep learning systems successfully solve many perception tasks such as object pose estimation when the input image is of high quality. However, in challenging imaging conditions such as on low resolution images or when the image is corrupted by imaging artifacts, current systems degrade considerably in accuracy. While a loss in performance is unavoidable we would like our models to quantify their uncertainty in order to achieve robustness against images of varying quality. Probabilistic deep learning models combine the expressive power of deep learning with uncertainty quantification. In this paper, we propose a novel probabilistic deep learning model for the task of angular regression. Our model uses von Mises distributions to predict a distribution over object pose angle. Whereas a single von Mises distribution is making strong assumptions about the shape of the distribution, we extend the basic model to predict a mixture of von Mises distributions. We show how to learn a mixture model using a finite and infinite number of mixture components. Our model allow for likelihood-based training and efficient inference at test time. We demonstrate on a number of challenging pose estimation datasets that our model produces calibrated probability predictions and competitive or superior point estimates compared to the current state-of-the-art.

ps

code pdf [BibTex]

code pdf [BibTex]


Recovering Accurate {3D} Human Pose in The Wild Using {IMUs} and a Moving Camera
Recovering Accurate 3D Human Pose in The Wild Using IMUs and a Moving Camera

Marcard, T. V., Henschel, R., Black, M. J., Rosenhahn, B., Pons-Moll, G.

In European Conference on Computer Vision (ECCV), Lecture Notes in Computer Science, vol 11214, pages: 614-631, Springer, Cham, September 2018 (inproceedings)

Abstract
In this work, we propose a method that combines a single hand-held camera and a set of Inertial Measurement Units (IMUs) attached at the body limbs to estimate accurate 3D poses in the wild. This poses many new challenges: the moving camera, heading drift, cluttered background, occlusions and many people visible in the video. We associate 2D pose detections in each image to the corresponding IMU-equipped persons by solving a novel graph based optimization problem that forces 3D to 2D coherency within a frame and across long range frames. Given associations, we jointly optimize the pose of a statistical body model, the camera pose and heading drift using a continuous optimization framework. We validated our method on the TotalCapture dataset, which provides video and IMU synchronized with ground truth. We obtain an accuracy of 26mm, which makes it accurate enough to serve as a benchmark for image-based 3D pose estimation in the wild. Using our method, we recorded 3D Poses in the Wild (3DPW ), a new dataset consisting of more than 51; 000 frames with accurate 3D pose in challenging sequences, including walking in the city, going up-stairs, having co ffee or taking the bus. We make the reconstructed 3D poses, video, IMU and 3D models available for research purposes at http://virtualhumans.mpi-inf.mpg.de/3DPW.

ps

pdf SupMat data project DOI Project Page [BibTex]

pdf SupMat data project DOI Project Page [BibTex]


Decentralized {MPC} based Obstacle Avoidance for Multi-Robot Target Tracking Scenarios
Decentralized MPC based Obstacle Avoidance for Multi-Robot Target Tracking Scenarios

Tallamraju, R., Rajappa, S., Black, M. J., Karlapalem, K., Ahmad, A.

2018 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pages: 1-8, IEEE, August 2018 (conference)

Abstract
In this work, we consider the problem of decentralized multi-robot target tracking and obstacle avoidance in dynamic environments. Each robot executes a local motion planning algorithm which is based on model predictive control (MPC). The planner is designed as a quadratic program, subject to constraints on robot dynamics and obstacle avoidance. Repulsive potential field functions are employed to avoid obstacles. The novelty of our approach lies in embedding these non-linear potential field functions as constraints within a convex optimization framework. Our method convexifies nonconvex constraints and dependencies, by replacing them as pre-computed external input forces in robot dynamics. The proposed algorithm additionally incorporates different methods to avoid field local minima problems associated with using potential field functions in planning. The motion planner does not enforce predefined trajectories or any formation geometry on the robots and is a comprehensive solution for cooperative obstacle avoidance in the context of multi-robot target tracking. We perform simulation studies for different scenarios to showcase the convergence and efficacy of the proposed algorithm.

ps

Published Version link (url) DOI [BibTex]

Published Version link (url) DOI [BibTex]


no image
Kernel Recursive ABC: Point Estimation with Intractable Likelihood

Kajihara, T., Kanagawa, M., Yamazaki, K., Fukumizu, K.

Proceedings of the 35th International Conference on Machine Learning, pages: 2405-2414, PMLR, July 2018 (conference)

Abstract
We propose a novel approach to parameter estimation for simulator-based statistical models with intractable likelihood. Our proposed method involves recursive application of kernel ABC and kernel herding to the same observed data. We provide a theoretical explanation regarding why the approach works, showing (for the population setting) that, under a certain assumption, point estimates obtained with this method converge to the true parameter, as recursion proceeds. We have conducted a variety of numerical experiments, including parameter estimation for a real-world pedestrian flow simulator, and show that in most cases our method outperforms existing approaches.

pn

Paper [BibTex]

Paper [BibTex]


Probabilistic Recurrent State-Space Models
Probabilistic Recurrent State-Space Models

Doerr, A., Daniel, C., Schiegg, M., Nguyen-Tuong, D., Schaal, S., Toussaint, M., Trimpe, S.

In Proceedings of the International Conference on Machine Learning (ICML), International Conference on Machine Learning (ICML), July 2018 (inproceedings)

Abstract
State-space models (SSMs) are a highly expressive model class for learning patterns in time series data and for system identification. Deterministic versions of SSMs (e.g., LSTMs) proved extremely successful in modeling complex time-series data. Fully probabilistic SSMs, however, unfortunately often prove hard to train, even for smaller problems. To overcome this limitation, we propose a scalable initialization and training algorithm based on doubly stochastic variational inference and Gaussian processes. In the variational approximation we propose in contrast to related approaches to fully capture the latent state temporal correlations to allow for robust training.

am ics

arXiv pdf Project Page [BibTex]

arXiv pdf Project Page [BibTex]


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

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

Workshop on Machine Learning for Causal Inference, Counterfactual Prediction, and Autonomous Action (CausalML) at ICML, July 2018 (conference)

ei pn

[BibTex]

[BibTex]


Online Learning of a Memory for Learning Rates
Online Learning of a Memory for Learning Rates

(nominated for best paper award)

Meier, F., Kappler, D., Schaal, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) 2018, IEEE, International Conference on Robotics and Automation, May 2018, accepted (inproceedings)

Abstract
The promise of learning to learn for robotics rests on the hope that by extracting some information about the learning process itself we can speed up subsequent similar learning tasks. Here, we introduce a computationally efficient online meta-learning algorithm that builds and optimizes a memory model of the optimal learning rate landscape from previously observed gradient behaviors. While performing task specific optimization, this memory of learning rates predicts how to scale currently observed gradients. After applying the gradient scaling our meta-learner updates its internal memory based on the observed effect its prediction had. Our meta-learner can be combined with any gradient-based optimizer, learns on the fly and can be transferred to new optimization tasks. In our evaluations we show that our meta-learning algorithm speeds up learning of MNIST classification and a variety of learning control tasks, either in batch or online learning settings.

am

pdf video code [BibTex]

pdf video code [BibTex]


Learning Sensor Feedback Models from Demonstrations via Phase-Modulated Neural Networks
Learning Sensor Feedback Models from Demonstrations via Phase-Modulated Neural Networks

Sutanto, G., Su, Z., Schaal, S., Meier, F.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) 2018, IEEE, International Conference on Robotics and Automation, May 2018 (inproceedings)

am

pdf video [BibTex]

pdf video [BibTex]


Co-Registration -- Simultaneous Alignment and Modeling of Articulated {3D} Shapes
Co-Registration – Simultaneous Alignment and Modeling of Articulated 3D Shapes

Black, M., Hirshberg, D., Loper, M., Rachlin, E., Weiss, A.

Febuary 2018, U.S.~Patent 9,898,848 (misc)

Abstract
Present application refers to a method, a model generation unit and a computer program (product) for generating trained models (M) of moving persons, based on physically measured person scan data (S). The approach is based on a common template (T) for the respective person and on the measured person scan data (S) in different shapes and different poses. Scan data are measured with a 3D laser scanner. A generic personal model is used for co-registering a set of person scan data (S) aligning the template (T) to the set of person scans (S) while simultaneously training the generic personal model to become a trained person model (M) by constraining the generic person model to be scan-specific, person-specific and pose-specific and providing the trained model (M), based on the co registering of the measured object scan data (S).

ps

text [BibTex]


End-to-end Recovery of Human Shape and Pose
End-to-end Recovery of Human Shape and Pose

Kanazawa, A., Black, M. J., Jacobs, D. W., Malik, J.

In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages: 7122-7131, IEEE Computer Society, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2018, 2018 (inproceedings)

Abstract
We describe Human Mesh Recovery (HMR), an end-to-end framework for reconstructing a full 3D mesh of a human body from a single RGB image. In contrast to most current methods that compute 2D or 3D joint locations, we produce a richer and more useful mesh representation that is parameterized by shape and 3D joint angles. The main objective is to minimize the reprojection loss of keypoints, which allows our model to be trained using in-the-wild images that only have ground truth 2D annotations. However, the reprojection loss alone is highly underconstrained. In this work we address this problem by introducing an adversary trained to tell whether human body shape and pose parameters are real or not using a large database of 3D human meshes. We show that HMR can be trained with and without using any paired 2D-to-3D supervision. We do not rely on intermediate 2D keypoint detections and infer 3D pose and shape parameters directly from image pixels. Our model runs in real-time given a bounding box containing the person. We demonstrate our approach on various images in-the-wild and out-perform previous optimization-based methods that output 3D meshes and show competitive results on tasks such as 3D joint location estimation and part segmentation.

ps

pdf code project video Project Page [BibTex]

pdf code project video Project Page [BibTex]


no image
On Time Optimization of Centroidal Momentum Dynamics

Ponton, B., Herzog, A., Del Prete, A., Schaal, S., Righetti, L.

In 2018 IEEE International Conference on Robotics and Automation (ICRA), pages: 5776-5782, IEEE, Brisbane, Australia, 2018 (inproceedings)

Abstract
Recently, the centroidal momentum dynamics has received substantial attention to plan dynamically consistent motions for robots with arms and legs in multi-contact scenarios. However, it is also non convex which renders any optimization approach difficult and timing is usually kept fixed in most trajectory optimization techniques to not introduce additional non convexities to the problem. But this can limit the versatility of the algorithms. In our previous work, we proposed a convex relaxation of the problem that allowed to efficiently compute momentum trajectories and contact forces. However, our approach could not minimize a desired angular momentum objective which seriously limited its applicability. Noticing that the non-convexity introduced by the time variables is of similar nature as the centroidal dynamics one, we propose two convex relaxations to the problem based on trust regions and soft constraints. The resulting approaches can compute time-optimized dynamically consistent trajectories sufficiently fast to make the approach realtime capable. The performance of the algorithm is demonstrated in several multi-contact scenarios for a humanoid robot. In particular, we show that the proposed convex relaxation of the original problem finds solutions that are consistent with the original non-convex problem and illustrate how timing optimization allows to find motion plans that would be difficult to plan with fixed timing † †Implementation details and demos can be found in the source code available at https://git-amd.tuebingen.mpg.de/bponton/timeoptimization.

am mg

link (url) DOI [BibTex]

link (url) DOI [BibTex]


Dissecting Adam: The Sign, Magnitude and Variance of Stochastic Gradients
Dissecting Adam: The Sign, Magnitude and Variance of Stochastic Gradients

Balles, L., Hennig, P.

In Proceedings of the 35th International Conference on Machine Learning (ICML), 2018 (inproceedings) Accepted

Abstract
The ADAM optimizer is exceedingly popular in the deep learning community. Often it works very well, sometimes it doesn't. Why? We interpret ADAM as a combination of two aspects: for each weight, the update direction is determined by the sign of stochastic gradients, whereas the update magnitude is determined by an estimate of their relative variance. We disentangle these two aspects and analyze them in isolation, gaining insight into the mechanisms underlying ADAM. This analysis also extends recent results on adverse effects of ADAM on generalization, isolating the sign aspect as the problematic one. Transferring the variance adaptation to SGD gives rise to a novel method, completing the practitioner's toolbox for problems where ADAM fails.

pn

link (url) Project Page [BibTex]

link (url) Project Page [BibTex]


no image
L4: Practical loss-based stepsize adaptation for deep learning

Rolinek, M., Martius, G.

In Advances in Neural Information Processing Systems 31 (NeurIPS 2018), pages: 6434-6444, (Editors: S. Bengio and H. Wallach and H. Larochelle and K. Grauman and N. Cesa-Bianchi and R. Garnett), Curran Associates, Inc., 2018 (inproceedings)

al

Github link (url) Project Page [BibTex]

Github link (url) Project Page [BibTex]


Systematic self-exploration of behaviors for robots in a dynamical systems framework
Systematic self-exploration of behaviors for robots in a dynamical systems framework

Pinneri, C., Martius, G.

In Proc. Artificial Life XI, pages: 319-326, MIT Press, Cambridge, MA, 2018 (inproceedings)

Abstract
One of the challenges of this century is to understand the neural mechanisms behind cognitive control and learning. Recent investigations propose biologically plausible synaptic mechanisms for self-organizing controllers, in the spirit of Hebbian learning. In particular, differential extrinsic plasticity (DEP) [Der and Martius, PNAS 2015], has proven to enable embodied agents to self-organize their individual sensorimotor development, and generate highly coordinated behaviors during their interaction with the environment. These behaviors are attractors of a dynamical system. In this paper, we use the DEP rule to generate attractors and we combine it with a “repelling potential” which allows the system to actively explore all its attractor behaviors in a systematic way. With a view to a self-determined exploration of goal-free behaviors, our framework enables switching between different motion patterns in an autonomous and sequential fashion. Our algorithm is able to recover all the attractor behaviors in a toy system and it is also effective in two simulated environments. A spherical robot discovers all its major rolling modes and a hexapod robot learns to locomote in 50 different ways in 30min.

al

link (url) DOI Project Page [BibTex]

link (url) DOI Project Page [BibTex]


Lions and Tigers and Bears: Capturing Non-Rigid, {3D}, Articulated Shape from Images
Lions and Tigers and Bears: Capturing Non-Rigid, 3D, Articulated Shape from Images

Zuffi, S., Kanazawa, A., Black, M. J.

In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE Computer Society, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2018, 2018 (inproceedings)

Abstract
Animals are widespread in nature and the analysis of their shape and motion is important in many fields and industries. Modeling 3D animal shape, however, is difficult because the 3D scanning methods used to capture human shape are not applicable to wild animals or natural settings. Consequently, we propose a method to capture the detailed 3D shape of animals from images alone. The articulated and deformable nature of animals makes this problem extremely challenging, particularly in unconstrained environments with moving and uncalibrated cameras. To make this possible, we use a strong prior model of articulated animal shape that we fit to the image data. We then deform the animal shape in a canonical reference pose such that it matches image evidence when articulated and projected into multiple images. Our method extracts significantly more 3D shape detail than previous methods and is able to model new species, including the shape of an extinct animal, using only a few video frames. Additionally, the projected 3D shapes are accurate enough to facilitate the extraction of a realistic texture map from multiple frames.

ps

pdf code/data 3D models Project Page [BibTex]

pdf code/data 3D models Project Page [BibTex]


Learning equations for extrapolation and control
Learning equations for extrapolation and control

Sahoo, S. S., Lampert, C. H., Martius, G.

In Proc. 35th International Conference on Machine Learning, ICML 2018, Stockholm, Sweden, 2018, 80, pages: 4442-4450, http://proceedings.mlr.press/v80/sahoo18a/sahoo18a.pdf, (Editors: Dy, Jennifer and Krause, Andreas), PMLR, 2018 (inproceedings)

Abstract
We present an approach to identify concise equations from data using a shallow neural network approach. In contrast to ordinary black-box regression, this approach allows understanding functional relations and generalizing them from observed data to unseen parts of the parameter space. We show how to extend the class of learnable equations for a recently proposed equation learning network to include divisions, and we improve the learning and model selection strategy to be useful for challenging real-world data. For systems governed by analytical expressions, our method can in many cases identify the true underlying equation and extrapolate to unseen domains. We demonstrate its effectiveness by experiments on a cart-pendulum system, where only 2 random rollouts are required to learn the forward dynamics and successfully achieve the swing-up task.

al

Code Arxiv Poster Slides link (url) Project Page [BibTex]

Code Arxiv Poster Slides link (url) Project Page [BibTex]


Robust Affordable 3D Haptic Sensation via Learning Deformation Patterns
Robust Affordable 3D Haptic Sensation via Learning Deformation Patterns

Sun, H., Martius, G.

Proceedings International Conference on Humanoid Robots, pages: 846-853, IEEE, New York, NY, USA, 2018 IEEE-RAS International Conference on Humanoid Robots, 2018, Oral Presentation (conference)

Abstract
Haptic sensation is an important modality for interacting with the real world. This paper proposes a general framework of inferring haptic forces on the surface of a 3D structure from internal deformations using a small number of physical sensors instead of employing dense sensor arrays. Using machine learning techniques, we optimize the sensor number and their placement and are able to obtain high-precision force inference for a robotic limb using as few as 9 sensors. For the optimal and sparse placement of the measurement units (strain gauges), we employ data-driven methods based on data obtained by finite element simulation. We compare data-driven approaches with model-based methods relying on geometric distance and information criteria such as Entropy and Mutual Information. We validate our approach on a modified limb of the “Poppy” robot [1] and obtain 8 mm localization precision.

al

DOI Project Page [BibTex]

DOI Project Page [BibTex]


no image
Unsupervised Contact Learning for Humanoid Estimation and Control

Rotella, N., Schaal, S., Righetti, L.

In 2018 IEEE International Conference on Robotics and Automation (ICRA), pages: 411-417, IEEE, Brisbane, Australia, 2018 (inproceedings)

Abstract
This work presents a method for contact state estimation using fuzzy clustering to learn contact probability for full, six-dimensional humanoid contacts. The data required for training is solely from proprioceptive sensors - endeffector contact wrench sensors and inertial measurement units (IMUs) - and the method is completely unsupervised. The resulting cluster means are used to efficiently compute the probability of contact in each of the six endeffector degrees of freedom (DoFs) independently. This clustering-based contact probability estimator is validated in a kinematics-based base state estimator in a simulation environment with realistic added sensor noise for locomotion over rough, low-friction terrain on which the robot is subject to foot slip and rotation. The proposed base state estimator which utilizes these six DoF contact probability estimates is shown to perform considerably better than that which determines kinematic contact constraints purely based on measured normal force.

am mg

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Learning Task-Specific Dynamics to Improve Whole-Body Control

Gams, A., Mason, S., Ude, A., Schaal, S., Righetti, L.

In Hua, IEEE, Beijing, China, November 2018 (inproceedings)

Abstract
In task-based inverse dynamics control, reference accelerations used to follow a desired plan can be broken down into feedforward and feedback trajectories. The feedback term accounts for tracking errors that are caused from inaccurate dynamic models or external disturbances. On underactuated, free-floating robots, such as humanoids, high feedback terms can be used to improve tracking accuracy; however, this can lead to very stiff behavior or poor tracking accuracy due to limited control bandwidth. In this paper, we show how to reduce the required contribution of the feedback controller by incorporating learned task-space reference accelerations. Thus, we i) improve the execution of the given specific task, and ii) offer the means to reduce feedback gains, providing for greater compliance of the system. With a systematic approach we also reduce heuristic tuning of the model parameters and feedback gains, often present in real-world experiments. In contrast to learning task-specific joint-torques, which might produce a similar effect but can lead to poor generalization, our approach directly learns the task-space dynamics of the center of mass of a humanoid robot. Simulated and real-world results on the lower part of the Sarcos Hermes humanoid robot demonstrate the applicability of the approach.

am mg

link (url) [BibTex]

link (url) [BibTex]


no image
An MPC Walking Framework With External Contact Forces

Mason, S., Rotella, N., Schaal, S., Righetti, L.

In 2018 IEEE International Conference on Robotics and Automation (ICRA), pages: 1785-1790, IEEE, Brisbane, Australia, May 2018 (inproceedings)

Abstract
In this work, we present an extension to a linear Model Predictive Control (MPC) scheme that plans external contact forces for the robot when given multiple contact locations and their corresponding friction cone. To this end, we set up a two-step optimization problem. In the first optimization, we compute the Center of Mass (CoM) trajectory, foot step locations, and introduce slack variables to account for violating the imposed constraints on the Zero Moment Point (ZMP). We then use the slack variables to trigger the second optimization, in which we calculate the optimal external force that compensates for the ZMP tracking error. This optimization considers multiple contacts positions within the environment by formulating the problem as a Mixed Integer Quadratic Program (MIQP) that can be solved at a speed between 100-300 Hz. Once contact is created, the MIQP reduces to a single Quadratic Program (QP) that can be solved in real-time ({\textless}; 1kHz). Simulations show that the presented walking control scheme can withstand disturbances 2-3× larger with the additional force provided by a hand contact.

am mg

link (url) DOI [BibTex]

link (url) DOI [BibTex]

2009


no image
Path integral-based stochastic optimal control for rigid body dynamics

Theodorou, E. A., Buchli, J., Schaal, S.

In Adaptive Dynamic Programming and Reinforcement Learning, 2009. ADPRL ’09. IEEE Symposium on, pages: 219-225, 2009, clmc (inproceedings)

Abstract
Recent advances on path integral stochastic optimal control [1],[2] provide new insights in the optimal control of nonlinear stochastic systems which are linear in the controls, with state independent and time invariant control transition matrix. Under these assumptions, the Hamilton-Jacobi-Bellman (HJB) equation is formulated and linearized with the use of the logarithmic transformation of the optimal value function. The resulting HJB is a linear second order partial differential equation which is solved by an approximation based on the Feynman-Kac formula [3]. In this work we review the theory of path integral control and derive the linearized HJB equation for systems with state dependent control transition matrix. In addition we derive the path integral formulation for the general class of systems with state dimensionality that is higher than the dimensionality of the controls. Furthermore, by means of a modified inverse dynamics controller, we apply path integral stochastic optimal control over the new control space. Simulations illustrate the theoretical results. Future developments and extensions are discussed.

am

link (url) [BibTex]

2009


link (url) [BibTex]


no image
Learning locomotion over rough terrain using terrain templates

Kalakrishnan, M., Buchli, J., Pastor, P., Schaal, S.

In Intelligent Robots and Systems, 2009. IROS 2009. IEEE/RSJ International Conference on, pages: 167-172, 2009, clmc (inproceedings)

Abstract
We address the problem of foothold selection in robotic legged locomotion over very rough terrain. The difficulty of the problem we address here is comparable to that of human rock-climbing, where foot/hand-hold selection is one of the most critical aspects. Previous work in this domain typically involves defining a reward function over footholds as a weighted linear combination of terrain features. However, a significant amount of effort needs to be spent in designing these features in order to model more complex decision functions, and hand-tuning their weights is not a trivial task. We propose the use of terrain templates, which are discretized height maps of the terrain under a foothold on different length scales, as an alternative to manually designed features. We describe an algorithm that can simultaneously learn a small set of templates and a foothold ranking function using these templates, from expert-demonstrated footholds. Using the LittleDog quadruped robot, we experimentally show that the use of terrain templates can produce complex ranking functions with higher performance than standard terrain features, and improved generalization to unseen terrain.

am

link (url) Project Page [BibTex]

link (url) Project Page [BibTex]


no image
CESAR: A lunar crater exploration and sample return robot

Schwendner, J., Grimminger, F., Bartsch, S., Kaupisch, T., Yüksel, M., Bresser, A., Akpo, J. B., Seydel, M. K. -., Dieterle, A., Schmidt, S., Kirchner, F.

In 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages: 3355-3360, October 2009 (inproceedings)

am

DOI [BibTex]

DOI [BibTex]


no image
Concept Evaluation of a New Biologically Inspired Robot “Littleape”

Kühn, D., Römmermann, M., Sauthoff, N., Grimminger, F., Kirchner, F.

In Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages: 589–594, IROS’09, IEEE Press, 2009 (inproceedings)

am

DOI [BibTex]

DOI [BibTex]


no image
Compact models of motor primitive variations for predictible reaching and obstacle avoidance

Stulp, F., Oztop, E., Pastor, P., Beetz, M., Schaal, S.

In IEEE-RAS International Conference on Humanoid Robots (Humanoids 2009), Paris, Dec.7-10, 2009, clmc (inproceedings)

Abstract
over and over again. This regularity allows humans and robots to reuse existing solutions for known recurring tasks. We expect that reusing a set of standard solutions to solve similar tasks will facilitate the design and on-line adaptation of the control systems of robots operating in human environments. In this paper, we derive a set of standard solutions for reaching behavior from human motion data. We also derive stereotypical reaching trajectories for variations of the task, in which obstacles are present. These stereotypical trajectories are then compactly represented with Dynamic Movement Primitives. On the humanoid robot Sarcos CB, this approach leads to reproducible, predictable, and human-like reaching motions.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Human optimization strategies under reward feedback

Hoffmann, H., Theodorou, E., Schaal, S.

In Abstracts of Neural Control of Movement Conference (NCM 2009), Waikoloa, Hawaii, 2009, 2009, clmc (inproceedings)

Abstract
Many hypothesis on human movement generation have been cast into an optimization framework, implying that movements are adapted to optimize a single quantity, like, e.g., jerk, end-point variance, or control cost. However, we still do not understand how humans actually learn when given only a cost or reward feedback at the end of a movement. Such a reinforcement learning setting has been extensively explored theoretically in engineering and computer science, but in human movement control, hardly any experiment studied movement learning under reward feedback. We present experiments probing which computational strategies humans use to optimize a movement under a continuous reward function. We present two experimental paradigms. The first paradigm mimics a ball-hitting task. Subjects (n=12) sat in front of a computer screen and moved a stylus on a tablet towards an unknown target. This target was located on a line that the subjects had to cross. During the movement, visual feedback was suppressed. After the movement, a reward was displayed graphically as a colored bar. As reward, we used a Gaussian function of the distance between the target location and the point of line crossing. We chose such a function since in sensorimotor tasks, the cost or loss function that humans seem to represent is close to an inverted Gaussian function (Koerding and Wolpert 2004). The second paradigm mimics pocket billiards. On the same experimental setup as above, the computer screen displayed a pocket (two bars), a white disk, and a green disk. The goal was to hit with the white disk the green disk (as in a billiard collision), such that the green disk moved into the pocket. Subjects (n=8) manipulated with the stylus the white disk to effectively choose start point and movement direction. Reward feedback was implicitly given as hitting or missing the pocket with the green disk. In both paradigms, subjects increased the average reward over trials. The surprising result was that in these experiments, humans seem to prefer a strategy that uses a reward-weighted average over previous movements instead of gradient ascent. The literature on reinforcement learning is dominated by gradient-ascent methods. However, our computer simulations and theoretical analysis revealed that reward-weighted averaging is the more robust choice given the amount of movement variance observed in humans. Apparently, humans choose an optimization strategy that is suitable for their own movement variance.

am

[BibTex]

[BibTex]


no image
Concept evaluation of a new biologically inspired robot “LittleApe”

Kühn, D., Römmermann, M., Sauthoff, N., Grimminger, F., Kirchner, F.

In 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages: 589-594, October 2009 (inproceedings)

am

DOI [BibTex]

DOI [BibTex]


no image
Proprioceptive control of a hybrid legged-wheeled robot

Eich, M., Grimminger, F., Kirchner, F.

In 2008 IEEE International Conference on Robotics and Biomimetics, pages: 774-779, February 2009 (inproceedings)

am

DOI [BibTex]

DOI [BibTex]


no image
Learning and generalization of motor skills by learning from demonstration

Pastor, P., Hoffmann, H., Asfour, T., Schaal, S.

In International Conference on Robotics and Automation (ICRA2009), Kobe, Japan, May 12-19, 2009, 2009, clmc (inproceedings)

Abstract
We provide a general approach for learning robotic motor skills from human demonstration. To represent an observed movement, a non-linear differential equation is learned such that it reproduces this movement. Based on this representation, we build a library of movements by labeling each recorded movement according to task and context (e.g., grasping, placing, and releasing). Our differential equation is formulated such that generalization can be achieved simply by adapting a start and a goal parameter in the equation to the desired position values of a movement. For object manipulation, we present how our framework extends to the control of gripper orientation and finger position. The feasibility of our approach is demonstrated in simulation as well as on a real robot. The robot learned a pick-and-place operation and a water-serving task and could generalize these tasks to novel situations.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Compliant quadruped locomotion over rough terrain

Buchli, J., Kalakrishnan, M., Mistry, M., Pastor, P., Schaal, S.

In Intelligent Robots and Systems, 2009. IROS 2009. IEEE/RSJ International Conference on, pages: 814-820, 2009, clmc (inproceedings)

Abstract
Many critical elements for statically stable walking for legged robots have been known for a long time, including stability criteria based on support polygons, good foothold selection, recovery strategies to name a few. All these criteria have to be accounted for in the planning as well as the control phase. Most legged robots usually employ high gain position control, which means that it is crucially important that the planned reference trajectories are a good match for the actual terrain, and that tracking is accurate. Such an approach leads to conservative controllers, i.e. relatively low speed, ground speed matching, etc. Not surprisingly such controllers are not very robust - they are not suited for the real world use outside of the laboratory where the knowledge of the world is limited and error prone. Thus, to achieve robust robotic locomotion in the archetypical domain of legged systems, namely complex rough terrain, where the size of the obstacles are in the order of leg length, additional elements are required. A possible solution to improve the robustness of legged locomotion is to maximize the compliance of the controller. While compliance is trivially achieved by reduced feedback gains, for terrain requiring precise foot placement (e.g. climbing rocks, walking over pegs or cracks) compliance cannot be introduced at the cost of inferior tracking. Thus, model-based control and - in contrast to passive dynamic walkers - active balance control is required. To achieve these objectives, in this paper we add two crucial elements to legged locomotion, i.e., floating-base inverse dynamics control and predictive force control, and we show that these elements increase robustness in face of unknown and unanticipated perturbations (e.g. obstacles). Furthermore, we introduce a novel line-based COG trajectory planner, which yields a simpler algorithm than traditional polygon based methods and creates the appropriate input to our control system.We show results from bot- h simulation and real world of a robotic dog walking over non-perceived obstacles and rocky terrain. The results prove the effectivity of the inverse dynamics/force controller. The presented results show that we have all elements needed for robust all-terrain locomotion, which should also generalize to other legged systems, e.g., humanoid robots.

am

link (url) [BibTex]

link (url) [BibTex]


no image
Inertial parameter estimation of floating-base humanoid systems using partial force sensing

Mistry, M., Schaal, S., Yamane, K.

In IEEE-RAS International Conference on Humanoid Robots (Humanoids 2009), Paris, Dec.7-10, 2009, clmc (inproceedings)

Abstract
Recently, several controllers have been proposed for humanoid robots which rely on full-body dynamic models. The estimation of inertial parameters from data is a critical component for obtaining accurate models for control. However, floating base systems, such as humanoid robots, incur added challenges to this task (e.g. contact forces must be measured, contact states can change, etc.) In this work, we outline a theoretical framework for whole body inertial parameter estimation, including the unactuated floating base. Using a least squares minimization approach, conducted within the nullspace of unmeasured degrees of freedom, we are able to use a partial force sensor set for full-body estimation, e.g. using only joint torque sensors, allowing for estimation when contact force measurement is unavailable or unreliable (e.g. due to slipping, rolling contacts, etc.). We also propose how to determine the theoretical minimum force sensor set for full body estimation, and discuss the practical limitations of doing so.

am

link (url) [BibTex]

link (url) [BibTex]

2007


no image
Towards Machine Learning of Motor Skills

Peters, J., Schaal, S., Schölkopf, B.

In Proceedings of Autonome Mobile Systeme (AMS), pages: 138-144, (Editors: K Berns and T Luksch), 2007, clmc (inproceedings)

Abstract
Autonomous robots that can adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. Early approaches to this goal during the heydays of artificial intelligence research in the late 1980s, however, made it clear that an approach purely based on reasoning or human insights would not be able to model all the perceptuomotor tasks that a robot should fulfill. Instead, new hope was put in the growing wake of machine learning that promised fully adaptive control algorithms which learn both by observation and trial-and-error. However, to date, learning techniques have yet to fulfill this promise as only few methods manage to scale into the high-dimensional domains of manipulator robotics, or even the new upcoming trend of humanoid robotics, and usually scaling was only achieved in precisely pre-structured domains. In this paper, we investigate the ingredients for a general approach to motor skill learning in order to get one step closer towards human-like performance. For doing so, we study two ma jor components for such an approach, i.e., firstly, a theoretically well-founded general approach to representing the required control structures for task representation and execution and, secondly, appropriate learning algorithms which can be applied in this setting.

am ei

PDF DOI [BibTex]

2007


PDF DOI [BibTex]


no image
Reinforcement Learning for Optimal Control of Arm Movements

Theodorou, E., Peters, J., Schaal, S.

In Abstracts of the 37st Meeting of the Society of Neuroscience., Neuroscience, 2007, clmc (inproceedings)

Abstract
Every day motor behavior consists of a plethora of challenging motor skills from discrete movements such as reaching and throwing to rhythmic movements such as walking, drumming and running. How this plethora of motor skills can be learned remains an open question. In particular, is there any unifying computa-tional framework that could model the learning process of this variety of motor behaviors and at the same time be biologically plausible? In this work we aim to give an answer to these questions by providing a computational framework that unifies the learning mechanism of both rhythmic and discrete movements under optimization criteria, i.e., in a non-supervised trial-and-error fashion. Our suggested framework is based on Reinforcement Learning, which is mostly considered as too costly to be a plausible mechanism for learning com-plex limb movement. However, recent work on reinforcement learning with pol-icy gradients combined with parameterized movement primitives allows novel and more efficient algorithms. By using the representational power of such mo-tor primitives we show how rhythmic motor behaviors such as walking, squash-ing and drumming as well as discrete behaviors like reaching and grasping can be learned with biologically plausible algorithms. Using extensive simulations and by using different reward functions we provide results that support the hy-pothesis that Reinforcement Learning could be a viable candidate for motor learning of human motor behavior when other learning methods like supervised learning are not feasible.

am ei

[BibTex]

[BibTex]


no image
Reinforcement learning by reward-weighted regression for operational space control

Peters, J., Schaal, S.

In Proceedings of the 24th Annual International Conference on Machine Learning, pages: 745-750, ICML, 2007, clmc (inproceedings)

Abstract
Many robot control problems of practical importance, including operational space control, can be reformulated as immediate reward reinforcement learning problems. However, few of the known optimization or reinforcement learning algorithms can be used in online learning control for robots, as they are either prohibitively slow, do not scale to interesting domains of complex robots, or require trying out policies generated by random search, which are infeasible for a physical system. Using a generalization of the EM-base reinforcement learning framework suggested by Dayan & Hinton, we reduce the problem of learning with immediate rewards to a reward-weighted regression problem with an adaptive, integrated reward transformation for faster convergence. The resulting algorithm is efficient, learns smoothly without dangerous jumps in solution space, and works well in applications of complex high degree-of-freedom robots.

am ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Policy gradient methods for machine learning

Peters, J., Theodorou, E., Schaal, S.

In Proceedings of the 14th INFORMS Conference of the Applied Probability Society, pages: 97-98, Eindhoven, Netherlands, July 9-11, 2007, 2007, clmc (inproceedings)

Abstract
We present an in-depth survey of policy gradient methods as they are used in the machine learning community for optimizing parameterized, stochastic control policies in Markovian systems with respect to the expected reward. Despite having been developed separately in the reinforcement learning literature, policy gradient methods employ likelihood ratio gradient estimators as also suggested in the stochastic simulation optimization community. It is well-known that this approach to policy gradient estimation traditionally suffers from three drawbacks, i.e., large variance, a strong dependence on baseline functions and a inefficient gradient descent. In this talk, we will present a series of recent results which tackles each of these problems. The variance of the gradient estimation can be reduced significantly through recently introduced techniques such as optimal baselines, compatible function approximations and all-action gradients. However, as even the analytically obtainable policy gradients perform unnaturally slow, it required the step from ÔvanillaÕ policy gradient methods towards natural policy gradients in order to overcome the inefficiency of the gradient descent. This development resulted into the Natural Actor-Critic architecture which can be shown to be very efficient in application to motor primitive learning for robotics.

am ei

[BibTex]

[BibTex]


no image
Policy Learning for Motor Skills

Peters, J., Schaal, S.

In Proceedings of 14th International Conference on Neural Information Processing (ICONIP), pages: 233-242, (Editors: Ishikawa, M. , K. Doya, H. Miyamoto, T. Yamakawa), 2007, clmc (inproceedings)

Abstract
Policy learning which allows autonomous robots to adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. However, to date, learning techniques have yet to fulfill this promise as only few methods manage to scale into the high-dimensional domains of manipulator robotics, or even the new upcoming trend of humanoid robotics, and usually scaling was only achieved in precisely pre-structured domains. In this paper, we investigate the ingredients for a general approach policy learning with the goal of an application to motor skill refinement in order to get one step closer towards human-like performance. For doing so, we study two major components for such an approach, i.e., firstly, we study policy learning algorithms which can be applied in the general setting of motor skill learning, and, secondly, we study a theoretically well-founded general approach to representing the required control structures for task representation and execution.

am ei

PDF DOI [BibTex]

PDF DOI [BibTex]


no image
Reinforcement learning for operational space control

Peters, J., Schaal, S.

In Proceedings of the 2007 IEEE International Conference on Robotics and Automation, pages: 2111-2116, IEEE Computer Society, ICRA, 2007, clmc (inproceedings)

Abstract
While operational space control is of essential importance for robotics and well-understood from an analytical point of view, it can be prohibitively hard to achieve accurate control in face of modeling errors, which are inevitable in complex robots, e.g., humanoid robots. In such cases, learning control methods can offer an interesting alternative to analytical control algorithms. However, the resulting supervised learning problem is ill-defined as it requires to learn an inverse mapping of a usually redundant system, which is well known to suffer from the property of non-convexity of the solution space, i.e., the learning system could generate motor commands that try to steer the robot into physically impossible configurations. The important insight that many operational space control algorithms can be reformulated as optimal control problems, however, allows addressing this inverse learning problem in the framework of reinforcement learning. However, few of the known optimization or reinforcement learning algorithms can be used in online learning control for robots, as they are either prohibitively slow, do not scale to interesting domains of complex robots, or require trying out policies generated by random search, which are infeasible for a physical system. Using a generalization of the EM-based reinforcement learning framework suggested by Dayan & Hinton, we reduce the problem of learning with immediate rewards to a reward-weighted regression problem with an adaptive, integrated reward transformation for faster convergence. The resulting algorithm is efficient, learns smoothly without dangerous jumps in solution space, and works well in applications of complex high degree-of-freedom robots.

am ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Using reward-weighted regression for reinforcement learning of task space control

Peters, J., Schaal, S.

In Proceedings of the 2007 IEEE International Symposium on Approximate Dynamic Programming and Reinforcement Learning, pages: 262-267, Honolulu, Hawaii, April 1-5, 2007, 2007, clmc (inproceedings)

Abstract
In this paper, we evaluate different versions from the three main kinds of model-free policy gradient methods, i.e., finite difference gradients, `vanilla' policy gradients and natural policy gradients. Each of these methods is first presented in its simple form and subsequently refined and optimized. By carrying out numerous experiments on the cart pole regulator benchmark we aim to provide a useful baseline for future research on parameterized policy search algorithms. Portable C++ code is provided for both plant and algorithms; thus, the results in this paper can be reevaluated, reused and new algorithms can be inserted with ease.

am ei

link (url) DOI [BibTex]

link (url) DOI [BibTex]