Header logo is ei


2007


no image
Machine Learning of Motor Skills for Robotics

Peters, J.

University of Southern California, Los Angeles, CA, USA, University of Southern California, Los Angeles, CA, USA, 2007, clmc (phdthesis)

Abstract
Autonomous robots that can assist humans in situations of daily life have been a long standing vision of robotics, artificial intelligence, and cognitive sciences. A first step towards this goal is to create robots that can accomplish a multitude of different tasks, triggered by environmental context or higher level instruction. 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 and 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 thesis, 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 major 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. As a theoretical foundation, we first study a general framework to generate control laws for real robots with a particular focus on skills represented as dynamical systems in differential constraint form. We present a point-wise optimal control framework resulting from a generalization of Gauss' principle and show how various well-known robot control laws can be derived by modifying the metric of the employed cost function. The framework has been successfully applied to task space tracking control for holonomic systems for several different metrics on the anthropomorphic SARCOS Master Arm. In order to overcome the limiting requirement of accurate robot models, we first employ learning methods to find learning controllers for task space control. However, when learning to execute a redundant control problem, we face the general problem of the non-convexity of the solution space which can force the robot to steer into physically impossible configurations if supervised learning methods are employed without further consideration. This problem can be resolved using two major insights, i.e., the learning problem can be treated as locally convex and the cost function of the analytical framework can be used to ensure global consistency. Thus, we derive an immediate reinforcement learning algorithm from the expectation-maximization point of view which leads to a reward-weighted regression technique. This method can be used both for operational space control as well as general immediate reward reinforcement learning problems. We demonstrate the feasibility of the resulting framework on the problem of redundant end-effector tracking for both a simulated 3 degrees of freedom robot arm as well as for a simulated anthropomorphic SARCOS Master Arm. While learning to execute tasks in task space is an essential component to a general framework to motor skill learning, learning the actual task is of even higher importance, particularly as this issue is more frequently beyond the abilities of analytical approaches than execution. We focus on the learning of elemental tasks which can serve as the "building blocks of movement generation", called motor primitives. Motor primitives are parameterized task representations based on splines or nonlinear differential equations with desired attractor properties. While imitation learning of parameterized motor primitives is a relatively well-understood problem, the self-improvement by interaction of the system with the environment remains a challenging problem, tackled in the fourth chapter of this thesis. For pursuing this goal, we highlight the difficulties with current reinforcement learning methods, and outline both established and novel algorithms for the gradient-based improvement of parameterized policies. We compare these algorithms in the context of motor primitive learning, and show that our most modern algorithm, the Episodic Natural Actor-Critic outperforms previous algorithms by at least an order of magnitude. We demonstrate the efficiency of this reinforcement learning method in the application of learning to hit a baseball with an anthropomorphic robot arm. In conclusion, in this thesis, we have contributed a general framework for analytically computing robot control laws which can be used for deriving various previous control approaches and serves as foundation as well as inspiration for our learning algorithms. We have introduced two classes of novel reinforcement learning methods, i.e., the Natural Actor-Critic and the Reward-Weighted Regression algorithm. These algorithms have been used in order to replace the analytical components of the theoretical framework by learned representations. Evaluations have been performed on both simulated and real robot arms.

[BibTex]

2007

[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.

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.

[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.

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.

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Relative Entropy Policy Search

Peters, J.

CLMC Technical Report: TR-CLMC-2007-2, Computational Learning and Motor Control Lab, Los Angeles, CA, 2007, clmc (techreport)

Abstract
This technical report describes a cute idea of how to create new policy search approaches. It directly relates to the Natural Actor-Critic methods but allows the derivation of one shot solutions. Future work may include the application to interesting problems.

PDF link (url) [BibTex]

PDF link (url) [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.

link (url) DOI [BibTex]

link (url) DOI [BibTex]


no image
Evaluation of Policy Gradient Methods and Variants on the Cart-Pole Benchmark

Riedmiller, M., Peters, J., Schaal, S.

In Proceedings of the 2007 IEEE International Symposium on Approximate Dynamic Programming and Reinforcement Learning, pages: 254-261, ADPRL, 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.

PDF [BibTex]

PDF [BibTex]

2002


no image
Real-Time Statistical Learning for Oculomotor Control and Visuomotor Coordination

Vijayakumar, S., Souza, A., Peters, J., Conradt, J., Rutkowski, T., Ijspeert, A., Nakanishi, J., Inoue, M., Shibata, T., Wiryo, A., Itti, L., Amari, S., Schaal, S.

(Editors: Becker, S. , S. Thrun, K. Obermayer), Sixteenth Annual Conference on Neural Information Processing Systems (NIPS), December 2002 (poster)

Web [BibTex]

2002

Web [BibTex]


no image
Optimized Support Vector Machines for Nonstationary Signal Classification

Davy, M., Gretton, A., Doucet, A., Rayner, P.

IEEE Signal Processing Letters, 9(12):442-445, December 2002 (article)

Abstract
This letter describes an efficient method to perform nonstationary signal classification. A support vector machine (SVM) algorithm is introduced and its parameters optimised in a principled way. Simulations demonstrate that our low complexity method outperforms state-of-the-art nonstationary signal classification techniques.

PostScript Web DOI [BibTex]

PostScript Web DOI [BibTex]


no image
Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond

Schölkopf, B., Smola, A.

pages: 644, Adaptive Computation and Machine Learning, MIT Press, Cambridge, MA, USA, December 2002, Parts of this book, including an introduction to kernel methods, can be downloaded here. (book)

Abstract
In the 1990s, a new type of learning algorithm was developed, based on results from statistical learning theory: the Support Vector Machine (SVM). This gave rise to a new class of theoretically elegant learning machines that use a central concept of SVMs-kernels—for a number of learning tasks. Kernel machines provide a modular framework that can be adapted to different tasks and domains by the choice of the kernel function and the base algorithm. They are replacing neural networks in a variety of fields, including engineering, information retrieval, and bioinformatics. Learning with Kernels provides an introduction to SVMs and related kernel methods. Although the book begins with the basics, it also includes the latest research. It provides all of the concepts necessary to enable a reader equipped with some basic mathematical knowledge to enter the world of machine learning using theoretically well-founded yet easy-to-use kernel algorithms and to understand and apply the powerful algorithms that have been developed over the last few years.

Web [BibTex]

Web [BibTex]


no image
Surface-slant-from-texture discrimination: Effects of slant level and texture type

Rosas, P., Wichmann, F., Wagemans, J.

Journal of Vision, 2(7):300, Second Annual Meeting of the Vision Sciences Society (VSS), November 2002 (poster)

Abstract
The problem of surface-slant-from-texture was studied psychophysically by measuring the performances of five human subjects in a slant-discrimination task with a number of different types of textures: uniform lattices, randomly displaced lattices, polka dots, Voronoi tessellations, orthogonal sinusoidal plaid patterns, fractal or 1/f noise, “coherent” noise and a “diffusion-based” texture (leopard skin-like). The results show: (1) Improving performance with larger slants for all textures. (2) A “non-symmetrical” performance around a particular slant characterized by a psychometric function that is steeper in the direction of the more slanted orientation. (3) For sufficiently large slants (66 deg) there are no major differences in performance between any of the different textures. (4) For slants at 26, 37 and 53 degrees, however, there are marked differences between the different textures. (5) The observed differences in performance across textures for slants up to 53 degrees are systematic within subjects, and nearly so across them. This allows a rank-order of textures to be formed according to their “helpfulness” — that is, how easy the discrimination task is when a particular texture is mapped on the surface. Polka dots tended to allow the best slant discrimination performance, noise patterns the worst up to the large slant of 66 degrees at which performance was almost independent of the particular texture chosen. Finally, our large number of 2AFC trials (approximately 2800 trials per texture across subjects) and associated tight confidence intervals may enable us to find out about which statistical properties of the textures could be responsible for surface-slant-from-texture estimation, with the ultimate goal of being able to predict observer performance for any arbitrary texture.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Modelling Contrast Transfer in Spatial Vision

Wichmann, F.

Journal of Vision, 2(10):7, Second Annual Meeting of the Vision Sciences Society (VSS), November 2002 (poster)

Abstract
Much of our information about spatial vision comes from detection experiments involving low-contrast stimuli. Contrast discrimination experiments provide one way to explore the visual system's response to stimuli of higher contrast, the results of which allow different models of contrast processing (e.g. energy versus gain-control models) to be critically assessed (Wichmann & Henning, 1999). Studies of detection and discrimination using pulse train stimuli in noise, on the other hand, make predictions about the number, position and properties of noise sources within the processing stream (Henning, Bird & Wichmann, 2002). Here I report modelling results combining data from both sinusoidal and pulse train experiments in and without noise to arrive at a more tightly constrained model of early spatial vision.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Gender Classification of Human Faces

Graf, A., Wichmann, F.

In Biologically Motivated Computer Vision, pages: 1-18, (Editors: Bülthoff, H. H., S.W. Lee, T. A. Poggio and C. Wallraven), Springer, Berlin, Germany, Second International Workshop on Biologically Motivated Computer Vision (BMCV), November 2002 (inproceedings)

Abstract
This paper addresses the issue of combining pre-processing methods—dimensionality reduction using Principal Component Analysis (PCA) and Locally Linear Embedding (LLE)—with Support Vector Machine (SVM) classification for a behaviorally important task in humans: gender classification. A processed version of the MPI head database is used as stimulus set. First, summary statistics of the head database are studied. Subsequently the optimal parameters for LLE and the SVM are sought heuristically. These values are then used to compare the original face database with its processed counterpart and to assess the behavior of a SVM with respect to changes in illumination and perspective of the face images. Overall, PCA was superior in classification performance and allowed linear separability.

PDF PDF DOI [BibTex]

PDF PDF DOI [BibTex]


no image
Insect-Inspired Estimation of Self-Motion

Franz, MO., Chahl, JS.

In Biologically Motivated Computer Vision, (2525):171-180, LNCS, (Editors: Bülthoff, H.H. , S.W. Lee, T.A. Poggio, C. Wallraven), Springer, Berlin, Germany, Second International Workshop on Biologically Motivated Computer Vision (BMCV), November 2002 (inproceedings)

Abstract
The tangential neurons in the fly brain are sensitive to the typical optic flow patterns generated during self-motion. In this study, we examine whether a simplified linear model of these neurons can be used to estimate self-motion from the optic flow. We present a theory for the construction of an optimal linear estimator incorporating prior knowledge about the environment. The optimal estimator is tested on a gantry carrying an omnidirectional vision sensor. The experiments show that the proposed approach leads to accurate and robust estimates of rotation rates, whereas translation estimates turn out to be less reliable.

PDF PDF DOI [BibTex]

PDF PDF DOI [BibTex]


no image
Pulse train detection and discrimination in pink noise

Henning, G., Wichmann, F., Bird, C.

Journal of Vision, 2(7):229, Second Annual Meeting of the Vision Sciences Society (VSS), November 2002 (poster)

Abstract
Much of our information about spatial vision comes from detection experiments involving low-contrast stimuli. Contrast discrimination experiments provide one way to explore the visual system's response to stimuli of higher contrast. We explored both detection and contrast discrimination performance with sinusoidal and "pulse-train" (or line) gratings. Both types of grating had a fundamental spatial frequency of 2.09-c/deg but the pulse-train, ideally, contains, in addition to its fundamental component, all the harmonics of the fundamental. Although the 2.09-c/deg pulse-train produced on the display was measured and shown to contain at least 8 harmonics at equal contrast, it was no more detectable than its most detectable component; no benefit from having additional information at the harmonics was measurable. The addition of broadband "pink" noise, designed to equalize the detectability of the components of the pulse train, made it about a factor of four more detectable than any of its components. However, in contrast-discrimination experiments, with an in-phase pedestal or masking grating of the same form and phase as the signal and 15% contrast, the noise did not improve the discrimination performance of the pulse train relative to that of its sinusoidal components. In contrast, a 2.09-c/deg "super train," constructed to have 8 equally detectable harmonics, was a factor of five more detectable than any of its components. We discuss the implications of these observations for models of early vision in particular the implications for possible sources of internal noise.

Web DOI [BibTex]

Web DOI [BibTex]


no image
A New Discriminative Kernel from Probabilistic Models

Tsuda, K., Kawanabe, M., Rätsch, G., Sonnenburg, S., Müller, K.

Neural Computation, 14(10):2397-2414, October 2002 (article)

PDF [BibTex]

PDF [BibTex]


no image
Combining sensory Information to Improve Visualization

Ernst, M., Banks, M., Wichmann, F., Maloney, L., Bülthoff, H.

In Proceedings of the Conference on Visualization ‘02 (VIS ‘02), pages: 571-574, (Editors: Moorhead, R. , M. Joy), IEEE, Piscataway, NJ, USA, IEEE Conference on Visualization (VIS '02), October 2002 (inproceedings)

Abstract
Seemingly effortlessly the human brain reconstructs the three-dimensional environment surrounding us from the light pattern striking the eyes. This seems to be true across almost all viewing and lighting conditions. One important factor for this apparent easiness is the redundancy of information provided by the sensory organs. For example, perspective distortions, shading, motion parallax, or the disparity between the two eyes' images are all, at least partly, redundant signals which provide us with information about the three-dimensional layout of the visual scene. Our brain uses all these different sensory signals and combines the available information into a coherent percept. In displays visualizing data, however, the information is often highly reduced and abstracted, which may lead to an altered perception and therefore a misinterpretation of the visualized data. In this panel we will discuss mechanisms involved in the combination of sensory information and their implications for simulations using computer displays, as well as problems resulting from current display technology such as cathode-ray tubes.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Sampling Techniques for Kernel Methods

Achlioptas, D., McSherry, F., Schölkopf, B.

In Advances in neural information processing systems 14 , pages: 335-342, (Editors: TG Dietterich and S Becker and Z Ghahramani), MIT Press, Cambridge, MA, USA, 15th Annual Neural Information Processing Systems Conference (NIPS), September 2002 (inproceedings)

Abstract
We propose randomized techniques for speeding up Kernel Principal Component Analysis on three levels: sampling and quantization of the Gram matrix in training, randomized rounding in evaluating the kernel expansions, and random projections in evaluating the kernel itself. In all three cases, we give sharp bounds on the accuracy of the obtained approximations.

PDF Web [BibTex]

PDF Web [BibTex]


no image
The Infinite Hidden Markov Model

Beal, MJ., Ghahramani, Z., Rasmussen, CE.

In Advances in Neural Information Processing Systems 14, pages: 577-584, (Editors: Dietterich, T.G. , S. Becker, Z. Ghahramani), MIT Press, Cambridge, MA, USA, Fifteenth Annual Neural Information Processing Systems Conference (NIPS), September 2002 (inproceedings)

Abstract
We show that it is possible to extend hidden Markov models to have a countably infinite number of hidden states. By using the theory of Dirichlet processes we can implicitly integrate out the infinitely many transition parameters, leaving only three hyperparameters which can be learned from data. These three hyperparameters define a hierarchical Dirichlet process capable of capturing a rich set of transition dynamics. The three hyperparameters control the time scale of the dynamics, the sparsity of the underlying state-transition matrix, and the expected number of distinct hidden states in a finite sequence. In this framework it is also natural to allow the alphabet of emitted symbols to be infinite - consider, for example, symbols being possible words appearing in English text.

PDF Web [BibTex]

PDF Web [BibTex]


no image
A new discriminative kernel from probabilistic models

Tsuda, K., Kawanabe, M., Rätsch, G., Sonnenburg, S., Müller, K.

In Advances in Neural Information Processing Systems 14, pages: 977-984, (Editors: Dietterich, T.G. , S. Becker, Z. Ghahramani), MIT Press, Cambridge, MA, USA, Fifteenth Annual Neural Information Processing Systems Conference (NIPS), September 2002 (inproceedings)

Abstract
Recently, Jaakkola and Haussler proposed a method for constructing kernel functions from probabilistic models. Their so called \Fisher kernel" has been combined with discriminative classi ers such as SVM and applied successfully in e.g. DNA and protein analysis. Whereas the Fisher kernel (FK) is calculated from the marginal log-likelihood, we propose the TOP kernel derived from Tangent vectors Of Posterior log-odds. Furthermore, we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing the TOP kernel. In experiments our new discriminative TOP kernel compares favorably to the Fisher kernel.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Incorporating Invariances in Non-Linear Support Vector Machines

Chapelle, O., Schölkopf, B.

In Advances in Neural Information Processing Systems 14, pages: 609-616, (Editors: TG Dietterich and S Becker and Z Ghahramani), MIT Press, Cambridge, MA, USA, 15th Annual Neural Information Processing Systems Conference (NIPS), September 2002 (inproceedings)

Abstract
The choice of an SVM kernel corresponds to the choice of a representation of the data in a feature space and, to improve performance, it should therefore incorporate prior knowledge such as known transformation invariances. We propose a technique which extends earlier work and aims at incorporating invariances in nonlinear kernels. We show on a digit recognition task that the proposed approach is superior to the Virtual Support Vector method, which previously had been the method of choice.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Functional Genomics of Osteoarthritis

Aigner, T., Bartnik, E., Zien, A., Zimmer, R.

Pharmacogenomics, 3(5):635-650, September 2002 (article)

Web [BibTex]

Web [BibTex]


no image
Kernel feature spaces and nonlinear blind source separation

Harmeling, S., Ziehe, A., Kawanabe, M., Müller, K.

In Advances in Neural Information Processing Systems 14, pages: 761-768, (Editors: Dietterich, T. G., S. Becker, Z. Ghahramani), MIT Press, Cambridge, MA, USA, Fifteenth Annual Neural Information Processing Systems Conference (NIPS), September 2002 (inproceedings)

Abstract
In kernel based learning the data is mapped to a kernel feature space of a dimension that corresponds to the number of training data points. In practice, however, the data forms a smaller submanifold in feature space, a fact that has been used e.g. by reduced set techniques for SVMs. We propose a new mathematical construction that permits to adapt to the intrinsic dimension and to find an orthonormal basis of this submanifold. In doing so, computations get much simpler and more important our theoretical framework allows to derive elegant kernelized blind source separation (BSS) algorithms for arbitrary invertible nonlinear mixings. Experiments demonstrate the good performance and high computational efficiency of our kTDSEP algorithm for the problem of nonlinear BSS.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Constructing Boosting algorithms from SVMs: an application to one-class classification.

Rätsch, G., Mika, S., Schölkopf, B., Müller, K.

IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(9):1184-1199, September 2002 (article)

Abstract
We show via an equivalence of mathematical programs that a support vector (SV) algorithm can be translated into an equivalent boosting-like algorithm and vice versa. We exemplify this translation procedure for a new algorithm—one-class leveraging—starting from the one-class support vector machine (1-SVM). This is a first step toward unsupervised learning in a boosting framework. Building on so-called barrier methods known from the theory of constrained optimization, it returns a function, written as a convex combination of base hypotheses, that characterizes whether a given test point is likely to have been generated from the distribution underlying the training data. Simulations on one-class classification problems demonstrate the usefulness of our approach.

DOI [BibTex]

DOI [BibTex]


no image
Kernel Dependency Estimation

Weston, J., Chapelle, O., Elisseeff, A., Schölkopf, B., Vapnik, V.

(98), Max Planck Institute for Biological Cybernetics, August 2002 (techreport)

Abstract
We consider the learning problem of finding a dependency between a general class of objects and another, possibly different, general class of objects. The objects can be for example: vectors, images, strings, trees or graphs. Such a task is made possible by employing similarity measures in both input and output spaces using kernel functions, thus embedding the objects into vector spaces. Output kernels also make it possible to encode prior information and/or invariances in the loss function in an elegant way. We experimentally validate our approach on several tasks: mapping strings to strings, pattern recognition, and reconstruction from partial images.

PDF [BibTex]

PDF [BibTex]


no image
Phase information in the recognition of natural images

Braun, D., Wichmann, F., Gegenfurtner, K.

Perception, 31(ECVP Abstract Supplement):133, 25th European Conference on Visual Perception, August 2002 (poster)

Abstract
Fourier phase plays an important role in determining global image structure. For example, when the phase spectrum of an image of a flower is swapped with that of a tank, we usually perceive a tank, even though the amplitude spectrum is still that of the flower. Similarly, when the phase spectrum of an image is randomly swapped across frequencies, that is its Fourier energy is randomly distributed over the image, the resulting image becomes impossible to recognise. Our goal was to evaluate the effect of phase manipulations in a quantitative manner. Subjects viewed two images of natural scenes, one of which contained an animal (the target) embedded in the background. The spectra of the images were manipulated by adding random phase noise at each frequency. The phase noise was the independent variable, uniformly distributed between 0° and ±180°. Subjects were remarkably resistant to phase noise. Even with ±120° noise, subjects were still 75% correct. The proportion of correct answers closely followed the correlation between original and noise-distorted images. Thus it appears as if it was not the global phase information per se that determines our percept of natural images, but rather the effect of phase on local image features.

Web [BibTex]

Web [BibTex]


no image
Algorithms for Learning Function Distinguishable Regular Languages

Fernau, H., Radl, A.

In Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, pages: 64-73, (Editors: Caelli, T. , A. Amin, R. P.W. Duin, M. Kamel, D. de Ridder), Springer, Berlin, Germany, Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, August 2002 (inproceedings)

Abstract
Function distinguishable languages were introduced as a new methodology of defining characterizable subclasses of the regular languages which are learnable from text. Here, we give details on the implementation and the analysis of the corresponding learning algorithms. We also discuss problems which might occur in practical applications.

PDF DOI [BibTex]

PDF DOI [BibTex]


no image
Co-Clustering of Biological Networks and Gene Expression Data

Hanisch, D., Zien, A., Zimmer, R., Lengauer, T.

Bioinformatics, (Suppl 1):145S-154S, 18, July 2002 (article)

Abstract
Motivation: Large scale gene expression data are often analysed by clustering genes based on gene expression data alone, though a priori knowledge in the form of biological networks is available. The use of this additional information promises to improve exploratory analysis considerably. Results: We propose constructing a distance function which combines information from expression data and biological networks. Based on this function, we compute a joint clustering of genes and vertices of the network. This general approach is elaborated for metabolic networks. We define a graph distance function on such networks and combine it with a correlation-based distance function for gene expression measurements. A hierarchical clustering and an associated statistical measure is computed to arrive at a reasonable number of clusters. Our method is validated using expression data of the yeast diauxic shift. The resulting clusters are easily interpretable in terms of the biochemical network and the gene expression data and suggest that our method is able to automatically identify processes that are relevant under the measured conditions.

Web [BibTex]

Web [BibTex]


no image
Global Geometry of SVM Classifiers

Zhou, D., Xiao, B., Zhou, H., Dai, R.

Max Planck Institute for Biological Cybernetics, Tübingen, Germany, June 2002 (techreport)

Abstract
We construct an geometry framework for any norm Support Vector Machine (SVM) classifiers. Within this framework, separating hyperplanes, dual descriptions and solutions of SVM classifiers are constructed by a purely geometric fashion. In contrast with the optimization theory used in SVM classifiers, we have no complicated computations any more. Each step in our theory is guided by elegant geometric intuitions.

PDF PostScript [BibTex]

PDF PostScript [BibTex]


no image
Computationally Efficient Face Detection

Romdhani, S., Torr, P., Schölkopf, B., Blake, A.

(MSR-TR-2002-69), Microsoft Research, June 2002 (techreport)

Web [BibTex]

Web [BibTex]


no image
Confidence measures for protein fold recognition

Sommer, I., Zien, A., von Ohsen, N., Zimmer, R., Lengauer, T.

Bioinformatics, 18(6):802-812, June 2002 (article)

[BibTex]

[BibTex]


no image
Decision Boundary Pattern Selection for Support Vector Machines

Shin, H., Cho, S.

In Proc. of the Korean Data Mining Conference, pages: 33-41, Korean Data Mining Conference, May 2002 (inproceedings)

[BibTex]

[BibTex]


no image
The contributions of color to recognition memory for natural scenes

Wichmann, F., Sharpe, L., Gegenfurtner, K.

Journal of Experimental Psychology: Learning, Memory and Cognition, 28(3):509-520, May 2002 (article)

Abstract
The authors used a recognition memory paradigm to assess the influence of color information on visual memory for images of natural scenes. Subjects performed 5-10% better for colored than for black-and-white images independent of exposure duration. Experiment 2 indicated little influence of contrast once the images were suprathreshold, and Experiment 3 revealed that performance worsened when images were presented in color and tested in black and white, or vice versa, leading to the conclusion that the surface property color is part of the memory representation. Experiments 4 and 5 exclude the possibility that the superior recognition memory for colored images results solely from attentional factors or saliency. Finally, the recognition memory advantage disappears for falsely colored images of natural scenes: The improvement in recognition memory depends on the color congruence of presented images with learned knowledge about the color gamut found within natural scenes. The results can be accounted for within a multiple memory systems framework.

PDF Web DOI [BibTex]

PDF Web DOI [BibTex]


no image
k-NN based Pattern Selection for Support Vector Classifiers

Shin, H., Cho, S.

In Proc. of the Korean Industrial Engineers Conference, pages: 645-651, Korean Industrial Engineers Conference, May 2002 (inproceedings)

[BibTex]

[BibTex]


no image
Microarrays: How Many Do You Need?

Zien, A., Fluck, J., Zimmer, R., Lengauer, T.

In RECOMB 2002, pages: 321-330, ACM Press, New York, NY, USA, Sixth Annual International Conference on Research in Computational Molecular Biology, April 2002 (inproceedings)

Abstract
We estimate the number of microarrays that is required in order to gain reliable results from a common type of study: the pairwise comparison of different classes of samples. Current knowlegde seems to suffice for the construction of models that are realistic with respect to searches for individual differentially expressed genes. Such models allow to investigate the dependence of the required number of samples on the relevant parameters: the biological variability of the samples within each class; the fold changes in expression; the detection sensitivity of the microarrays; and the acceptable error rates of the results. We supply experimentalists with general conclusions as well as a freely accessible Java applet at http://cartan.gmd.de/~zien/classsize/ for fine tuning simulations to their particular actualities. Since the situation can be assumed to be very similar for large scale proteomics and metabolomics studies, our methods and results might also apply there.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Detection and discrimination in pink noise

Wichmann, F., Henning, G.

5, pages: 100, 5. T{\"u}binger Wahrnehmungskonferenz (TWK), February 2002 (poster)

Abstract
Much of our information about early spatial vision comes from detection experiments involving low-contrast stimuli, which are not, perhaps, particularly "natural" stimuli. Contrast discrimination experiments provide one way to explore the visual system's response to stimuli of higher contrast whilst keeping the number of unknown parameters comparatively small. We explored both detection and contrast discrimination performance with sinusoidal and "pulse-train" (or line) gratings. Both types of grating had a fundamental spatial frequency of 2.09-c/deg but the pulse-train, ideally, contains, in addition to its fundamental component, all the harmonics of the fundamental. Although the 2.09-c/deg pulse-train produced on our display was measured using a high-performance digital camera (Photometrics) and shown to contain at least 8 harmonics at equal contrast, it was no more detectable than its most detectable component; no benefit from having additional information at the harmonics was measurable. The addition of broadband 1-D "pink" noise made it about a factor of four more detectable than any of its components. However, in contrast-discrimination experiments, with an in-phase pedestal or masking grating of the same form and phase as the signal and 15% contrast, the noise did not improve the discrimination performance of the pulse train relative to that of its sinusoidal components. We discuss the implications of these observations for models of early vision in particular the implications for possible sources of internal noise.

Web [BibTex]

Web [BibTex]


no image
Nonlinear Multivariate Analysis with Geodesic Kernels

Kuss, M.

Biologische Kybernetik, Technische Universität Berlin, February 2002 (diplomathesis)

GZIP [BibTex]

GZIP [BibTex]


no image
Pattern Selection for Support Vector Classifiers

Shin, H., Cho, S.

In Ideal 2002, pages: 97-103, (Editors: Yin, H. , N. Allinson, R. Freeman, J. Keane, S. Hubbard), Springer, Berlin, Germany, Third International Conference on Intelligent Data Engineering and Automated Learning, January 2002 (inproceedings)

Abstract
SVMs tend to take a very long time to train with a large data set. If "redundant" patterns are identified and deleted in pre-processing, the training time could be reduced significantly. We propose a k-nearest neighbors(k-NN) based pattern selection method. The method tries to select the patterns that are near the decision boundary and that are correctly labeled. The simulations over synthetic data sets showed promising results: (1) By converting a non-separable problem to a separable one, the search for an optimal error tolerance parameter became unnecessary. (2) SVM training time decreased by two orders of magnitude without any loss of accuracy. (3) The redundant SVs were substantially reduced.

PDF Web DOI [BibTex]

PDF Web DOI [BibTex]


no image
Kernel-based nonlinear blind source separation

Harmeling, S., Ziehe, A., Kawanabe, M., Müller, K.

EU-Project BLISS, January 2002 (techreport)

GZIP [BibTex]

GZIP [BibTex]


no image
Training invariant support vector machines

DeCoste, D., Schölkopf, B.

Machine Learning, 46(1-3):161-190, January 2002 (article)

Abstract
Practical experience has shown that in order to obtain the best possible performance, prior knowledge about invariances of a classification problem at hand ought to be incorporated into the training procedure. We describe and review all known methods for doing so in support vector machines, provide experimental results, and discuss their respective merits. One of the significant new results reported in this work is our recent achievement of the lowest reported test error on the well-known MNIST digit recognition benchmark task, with SVM training times that are also significantly faster than previous SVM methods.

PDF DOI [BibTex]

PDF DOI [BibTex]


no image
Model Selection for Small Sample Regression

Chapelle, O., Vapnik, V., Bengio, Y.

Machine Learning, 48(1-3):9-23, 2002 (article)

Abstract
Model selection is an important ingredient of many machine learning algorithms, in particular when the sample size in small, in order to strike the right trade-off between overfitting and underfitting. Previous classical results for linear regression are based on an asymptotic analysis. We present a new penalization method for performing model selection for regression that is appropriate even for small samples. Our penalization is based on an accurate estimator of the ratio of the expected training error and the expected generalization error, in terms of the expected eigenvalues of the input covariance matrix.

PostScript [BibTex]

PostScript [BibTex]


no image
The leave-one-out kernel

Tsuda, K., Kawanabe, M.

In Artificial Neural Networks -- ICANN 2002, 2415, pages: 727-732, LNCS, (Editors: Dorronsoro, J. R.), Artificial Neural Networks -- ICANN, 2002 (inproceedings)

PDF [BibTex]

PDF [BibTex]


no image
Concentration Inequalities and Empirical Processes Theory Applied to the Analysis of Learning Algorithms

Bousquet, O.

Biologische Kybernetik, Ecole Polytechnique, 2002 (phdthesis) Accepted

Abstract
New classification algorithms based on the notion of 'margin' (e.g. Support Vector Machines, Boosting) have recently been developed. The goal of this thesis is to better understand how they work, via a study of their theoretical performance. In order to do this, a general framework for real-valued classification is proposed. In this framework, it appears that the natural tools to use are Concentration Inequalities and Empirical Processes Theory. Thanks to an adaptation of these tools, a new measure of the size of a class of functions is introduced, which can be computed from the data. This allows, on the one hand, to better understand the role of eigenvalues of the kernel matrix in Support Vector Machines, and on the other hand, to obtain empirical model selection criteria.

PostScript [BibTex]


no image
Support Vector Machines: Induction Principle, Adaptive Tuning and Prior Knowledge

Chapelle, O.

Biologische Kybernetik, 2002 (phdthesis)

Abstract
This thesis presents a theoretical and practical study of Support Vector Machines (SVM) and related learning algorithms. In a first part, we introduce a new induction principle from which SVMs can be derived, but some new algorithms are also presented in this framework. In a second part, after studying how to estimate the generalization error of an SVM, we suggest to choose the kernel parameters of an SVM by minimizing this estimate. Several applications such as feature selection are presented. Finally the third part deals with the incoporation of prior knowledge in a learning algorithm and more specifically, we studied the case of known invariant transormations and the use of unlabeled data.

GZIP [BibTex]


no image
Contrast discrimination with sinusoidal gratings of different spatial frequency

Bird, C., Henning, G., Wichmann, F.

Journal of the Optical Society of America A, 19(7), pages: 1267-1273, 2002 (article)

Abstract
The detectability of contrast increments was measured as a function of the contrast of a masking or “pedestal” grating at a number of different spatial frequencies ranging from 2 to 16 cycles per degree of visual angle. The pedestal grating always had the same orientation, spatial frequency and phase as the signal. The shape of the contrast increment threshold versus pedestal contrast (TvC) functions depend of the performance level used to define the “threshold,” but when both axes are normalized by the contrast corresponding to 75% correct detection at each frequency, the (TvC) functions at a given performance level are identical. Confidence intervals on the slope of the rising part of the TvC functions are so wide that it is not possible with our data to reject Weber’s Law.

PDF [BibTex]

PDF [BibTex]


no image
A compression approach to support vector model selection

von Luxburg, U., Bousquet, O., Schölkopf, B.

(101), Max Planck Institute for Biological Cybernetics, 2002, see more detailed JMLR version (techreport)

Abstract
In this paper we investigate connections between statistical learning theory and data compression on the basis of support vector machine (SVM) model selection. Inspired by several generalization bounds we construct ``compression coefficients'' for SVMs, which measure the amount by which the training labels can be compressed by some classification hypothesis. The main idea is to relate the coding precision of this hypothesis to the width of the margin of the SVM. The compression coefficients connect well known quantities such as the radius-margin ratio R^2/rho^2, the eigenvalues of the kernel matrix and the number of support vectors. To test whether they are useful in practice we ran model selection experiments on several real world datasets. As a result we found that compression coefficients can fairly accurately predict the parameters for which the test error is minimized.

[BibTex]

[BibTex]


no image
A Bennett Concentration Inequality and Its Application to Suprema of Empirical Processes

Bousquet, O.

C. R. Acad. Sci. Paris, Ser. I, 334, pages: 495-500, 2002 (article)

Abstract
We introduce new concentration inequalities for functions on product spaces. They allow to obtain a Bennett type deviation bound for suprema of empirical processes indexed by upper bounded functions. The result is an improvement on Rio's version \cite{Rio01b} of Talagrand's inequality \cite{Talagrand96} for equidistributed variables.

PDF PostScript [BibTex]


no image
Numerical evolution of axisymmetric, isolated systems in general relativity

Frauendiener, J., Hein, M.

Physical Review D, 66, pages: 124004-124004, 2002 (article)

Abstract
We describe in this article a new code for evolving axisymmetric isolated systems in general relativity. Such systems are described by asymptotically flat space-times, which have the property that they admit a conformal extension. We are working directly in the extended conformal manifold and solve numerically Friedrich's conformal field equations, which state that Einstein's equations hold in the physical space-time. Because of the compactness of the conformal space-time the entire space-time can be calculated on a finite numerical grid. We describe in detail the numerical scheme, especially the treatment of the axisymmetry and the boundary.

GZIP [BibTex]

GZIP [BibTex]