Techniques of vision-based motion analysis aim to detect, track, identify, and generally understand the behavior of objects in image sequences. With the growth of video data in a wide range of applications from visual surveillance to human-machine interfaces, the ability to automatically analyze and understand object motions from video footage is of increasing importance. Among the latest developments in this field is the application of statistical machine learning algorithms for object tracking, activity modeling, and recognition.

Developed from expert contributions to the first and second International Workshop on Machine Learning for Vision-Based Motion Analysis, this important text/reference highlights the latest algorithms and systems for robust and effective vision-based motion understanding from a machine learning perspective. Highlighting the benefits of collaboration between the communities of object motion understanding and machine learning, the book discusses the most active forefronts of research, including current challenges and potential future directions.

Topics and features:

• Provides a comprehensive review of the latest developments in vision-based motion analysis, presenting numerous case studies on state-of-the-art learning algorithms
• Examines algorithms for clustering and segmentation, and manifold learning for dynamical models
• Describes the theory behind mixed-state statistical models, with a focus on mixed-state Markov models that take into account spatial and temporal interaction
• Discusses object tracking in surveillance image streams, discriminative multiple target tracking, and guidewire tracking in fluoroscopy
• Explores issues of modeling for saliency detection, human gait modeling, modeling of extremely crowded scenes, and behavior modeling from video surveillance data
• Investigates methods for automatic recognition of gestures in Sign Language, and human action recognition from small training sets

Researchers, professional engineers, and graduate students in computer vision, pattern recognition and machine learning, will all find this text an accessible survey of machine learning techniques for vision-based motion analysis. The book will also be of interest to all who work with specific vision applications, such as surveillance, sport event analysis, healthcare, video conferencing, and motion video indexing and retrieval.

With the exceptional increase in computing power, storage capacity and network bandwidth of the past decades, ever growing datasets are collected in fields such as bioinformatics (Splice Sites, Gene Boundaries, etc), IT-security (Network traffic) or Text-Classification (Spam vs. Non-Spam), to name but a few. While the data size growth leaves computational methods as the only viable way of dealing with data, it poses new challenges to ML methods.

This PASCAL Challenge is concerned with the scalability and efficiency of existing ML approaches with respect to computational, memory or communication resources, e.g. resulting from a high algorithmic complexity, from the size or dimensionality of the data set, and from the trade-off between distributed resolution and communication costs.

Indeed many comparisons are presented in the literature; however, these usually focus on assessing a few algorithms, or considering a few datasets; further, they most usually involve different evaluation criteria, model parameters and stopping conditions. As a result it is difficult to determine how does a method behave and compare with the other ones in terms of test error, training time and memory requirements, which are the practically relevant criteria.

We are therefore organizing a competition, that is designed to be fair and enables a direct comparison of current large scale classifiers aimed at answering the question “Which learning method is the most accurate given limited resources?”. To this end we provide a generic evaluation framework tailored to the specifics of the competing methods. Providing a wide range of datasets, each of which having specific properties we propose to evaluate the methods based on performance figures, displaying training time vs. test error, dataset size vs. test error and dataset size vs. training time.

Organizers

• Soeren Sonnenburg, TU Berlin, Berlin, Germany
• Vojtech Franc, Czech Technical University, Prague, Czech Republic
• Elad Yom-Tov, IBM Haifa Research Lab, Haifa, Israel
• Michele Sebag, LRI, Orsay, France

During the last few years (2004-2007), there have been several breakthroughs in the area of Minimum
Description Length (MDL) modeling, learning and prediction. These breakthroughs concern the efficient computation and proper formulation of MDL in parametric problems based on the “normalized maximum likelihood”, as well as altogether new, and better, coding schemes for nonparametric problems.

This essentially solves the so-called AIC-BIC dilemma, which has been a central problem in statistical model selection for more than 20 years now. The goal of this workshop is to introduce these exciting new developments to the ML and UAI communities, and to foster new collaborations between interested researchers.

Most new developments that are the focus of this workshop concern efficient (in many cases, linear- time) algorithms for theoretically optimal inference procedures that were previously thought not to be efficiently solvable. It is therefore hoped that the workshop will inspire original practical applications of MDL in machine learning domains.

Development of such applications recently became a lot easier, because of the new (2007) book on MDL by P. Grunwald, which provides the first comprehensive overview of the field, as well as in-depth discussions of how it relates to other approaches such as Bayesian inference. Remarkably, the originator of MDL, J. Rissanen, also published a new monograph in 2007;
and a Festschrift in Honor of Rissanen’s 75th birthday was presented to him in May 2008.

Organizing Committee

With the current explosion and quick expansion of music in digital formats,  research on machine learning and music is gaining increasing popularity. As complexity of the problems investigated by researchers on this area increases, there is a need to develop new algorithms and methods to solve these problems. Machine learning has proved to provide efficient solutions to many music-related problems both of academic and commercial interest.

MML 2008 intends to continue a series of events related to artificial intelligence/machine learning and music that have been held in the past. The presence of such events and the increasing number of contributions they receive indicates that the application of related techniques to the development of music processing systems is an active, exciting and significant area of research which has become an established field of research. The goal of the workshop is to bring together researchers who are using machine learning in musical applications, providing the opportunity to promote, present and discuss ongoing work in the area. MML 2007 is planned to last one full day, and will feature paper presentations, panel discussions and open discussions. Accepted contributions will be available from the workshop web page as soon as possible in order to encourage active discussion during the workshop.

Organizing Committee

• Rafael Ramirez
• Universitat Pompeu Fabra, Barcelona, Spain
• Christina Anagnostopoulou, University of Athens, Athens, Greece
• Darrell Conklin, City University, London, UK
• José Manuel Iñesta, Alicante University, Spain
• Xavier Serra,Universitat Pompeu Fabra, Barcelona, Spain

 Programme Committee

• Christina Anagnostopoulou (University of Athens, Greece)
• Darrell Conklin (City University, UK)
• Judy Franklin (Smith Collage, USA)
• Fabien Gouyon (INESC Porto, Portugal)
• Maarten Grachten (Johannes Kepler University, Austria)
• Perfecto Herrera (Universitat Pompeu Fabra, Spain)
• Potamitis Ilyas (Technological Educational Institute of Crete, Greece)
• José Manuel Iñesta  (Alicante University, Spain)
• Søren Madsen (Austrian Research Inst. for Artificial Intelligence, Austria)
• Pedro Ponce de León  (Alicante University, Spain)
• Aggelos Pikrakis  (University of Piraeus, Greece)
• Hendrik Purwins (Universitat Pompeu Fabra, Spain)
• Rafael Ramirez (Universitat Pompeu Fabra, Spain)

One of the major problems driving current research in statistical machine learning is the search for ways to exploit highly-structured models that are both expressive and tractable. Nonparametric Bayesian methodology provides significant leverage on this problem. In the nonparametric Bayesian framework, the prior distribution is not a fixed parametric form, but is rather a general stochastic process—-a distribution over a possibly uncountably infinite number of random variables. This generality makes it possible to work with prior and posterior distributions on objects such as trees of unbounded depth and breadth, graphs, partitions, sets of monotone functions, sets of smooth functions and sets of general measures.

Applications of nonparametric Bayesian methods have begun to appear in disciplines such as information retrieval, natural language processing, machine vision, computational biology, cognitive science and signal processing. Because of their flexibility, they can also be used to express prior knowledge without restricting to small parametric classes. Furthermore, research on nonparametric Bayesian models has served to enhance the links between statistical machine learning and a number of other mathematical disciplines, including stochastic processes, algorithms, optimization, combinatorics and knowledge representation.

There have been several previous workshops on nonparametric Bayesian methods at machine learning conferences, including workshops at NIPS in 2003 and 2005 and a workshop at ICML in 2006. This workshop aims to build on the success of these earlier workshops to catalyze further research. There are many problem areas that need additional attention, and this workshop is intended to bring together the growing community of nonparametric Bayesian researchers to explore these issues.

• Software The development of general software packages is not only of obvious practical significance but can also lead to important theoretical advances. Building nonparametric inference software packages will help us understand where we are faced with fundamental limits and where merely with engineering challenges, just as efforts towards general purpose Bayesian network algorithms led to the discovery of the importance of tree width. While most current algorithms are very model-specific, striving for general purpose methods will help bring about the theoretical framework to discuss these non-parametric models as a family, and the language to describe their various combinations. Last but not least, such software would allow a much larger community to reap the benefits of this research. In return, this field experience would quickly highlight the strengths and weaknesses of current methods, and draw attention to the most pressing needs.

• Bridging communities This field attracts researchers from a broad range of disciplines, ranging from theoretical statisticians and probabilists to people building very specialized applications. It is important that we effectively communicate our advances and needs to better focus our effort. Theoreticians need to know which methods are used in practice and why, and what common tricks seem to help, while applied researchers will want to hear about the latest models and inference algorithms. It is especially important to bring statisticians and machine learning researchers together, since these communities work on the closely related topics, but have complementary strengths, often use different terminology, and focus on different application areas.

• Efficient inference A key focus of software development, and a top concern of potential users, is scalability. Markov chain Monte Carlo methods have proved their versatility and various advances have greatly improved their speed, but ensuring and assessing convergence remains a difficult topic and it is still unclear whether they will be reliable enough for non experts to use them with confidence. Variational methods, where they have been applied, have brought great speedups and reliable convergence, often at little cost in accuracy, but designing these methods largely remains an art. Additionally, without a better understanding of the loss in accuracy incurred by this approximation, it is possible that this cost will increase in more complex models. This meeting will help us summarize what works, what doesn't, and why, and discuss how to assess performance and build benchmark datasets.

The main aim of this workshop is to allow leading Bayesian researchers in machine learning to get together presenting their latest ideas and discussing future directions. The workshop will provide a forum to discuss Bayesian inference in machine learning. A particular focus will be on how Bayesian inference can be used to encode prior knowledge.

Themes

• Incorporating Complex Prior Knowledge in Bayesian inference, for example mechanistic models (such as differential equations) or knowledge transfered from other related situations (e.g. hierarchical Dirichlet Processes).
• Model mismatch: the Bayesian lynch pin is that the model is correct, but it rarely is.
• Approximation techniques: how should we do Bayesian inference in practice. Sampling, variational, Laplace or something else?
• Your pet Bayesian issue here.

Supervised and Unsupervised Ensemble Methods and Their Applications (SUEMA 2008) is a workshop organized in the context of the European Conference on Artificial Intelligence (ECAI 2008), and within the PASCAL 2 (Pattern Analysis, Statistical Modelling and Computational Learning) European Network of Excellence.

This second edition, follows up the first one held in Girona (Spain) within the IbPRIA Conference, and intends to provide a forum for researchers in the field of Artificial Intelligence and Machine Learning to discuss topics related to ensemble methods and their applications.

New trends and challenges in the theory and methods of ensembles of learning machines will be discussed, with a particular focus on the application of principled ensemble methods to emerging real-life problems.

Time-series analysis throws up interesting problems which remain fundamental to several key and as yet unsolved application areas. For example, Bayesian time-series models typically couple all time-points of the series, resulting in intractable inference in high-dimensional latent spaces and therefore requiring approximation.

The workshop will discuss both theories and applications related to probabilistic approaches to time-series analysis. Viewpoints and experiences from researchers belonging to different communities, including machine learning, statistics and statistical physics, are particularly encouraged. For example, approximate inference in the machine learning community tends to be more focussed on deterministic/variational approaches, whilst the statistics community tends to prefer sampling approaches. Amongst others, discussions related to this topic would be appreciated.

More generally, advances in practical and theoretical issues related to probabilistic approaches to time-series modelling, including for example inference, estimation, prediction, classification, clustering and source separation, are appreciated. Novel application areas and the challenges that they bring are also welcome.

Inference and Estimation in Probabilistic Time-Series Models

Inference and Estimation in Probabilistic Time-Series Models

Inference and Estimation in Probabilistic Time-Series Models

Inference and Estimation in Probabilistic Time-Series Models

Inference and Estimation in Probabilistic Time-Series Models

Time-series analysis throws up interesting problems which remain fundamental to several key and as yet unsolved application areas. For example, Bayesian time-series models typically couple all time-points of the series, resulting in intractable inference in high-dimensional latent spaces and therefore requiring approximation.

The workshop will discuss both theories and applications related to probabilistic approaches to time-series analysis. Viewpoints and experiences from researchers belonging to different communities, including machine learning, statistics and statistical physics, are particularly encouraged. For example, approximate inference in the machine learning community tends to be more focussed on deterministic/variational approaches, whilst the statistics community tends to prefer sampling approaches. Amongst others, discussions related to this topic would be appreciated.

More generally, advances in practical and theoretical issues related to probabilistic approaches to time-series modelling, including for example inference, estimation, prediction, classification, clustering and source separation, are appreciated. Novel application areas and the challenges that they bring are also welcome.

Organisers

• David Barber (UCL London, UK)
• Silvia Chiappa (MPI Tuebingen, Germany)
• Taylan Cemgil (University of Cambridge, UK)