Our goal is to establish the Joint Workshop on Automatic Knowledge Base Construction and Web-scale Knowledge Extraction (AKBC-WEKEX 2012) as a venue of excellence and vision in the area of knowledge harvesting from text. With invited talks by leading researchers from industry, academia, and the government, and by focusing particularly on vision papers, we aim to provide a vivid forum of discussion about the field of automated knowledge base construction.

Joint Workshop on Automatic Knowledge Base Construction and Web-scale Knowledge Extraction Workshop

Joint Workshop on Automatic Knowledge Base Construction and Web-scale Knowledge Extraction Workshop

Joint Workshop on Automatic Knowledge Base Construction and Web-scale Knowledge Extraction Workshop

Processing in the brain in general and visual processing in particular is organized in a hierarchical fashion, from simple localized features to complex, large scale features. The visual system consists of a hierarchy, in which neurons in early visual areas extract simple image features (orientation, motion, disparity) over a small local region of visual space, which are then transmitted to neurons in higher visual areas responding to more complex features (e.g. shape) over a larger region of visual space. Hierarchical representations can derive and organize features at multiple levels. Hierarchical representations with many levels are called 'deep hierarchies'. They build on top of each other by exploiting the shareability of features among more complex compositions or objects themselves. Sharing features, on the one hand, means sharing common computations, which brings about (highly desirable) computational efficiency. On the other hand, reusing the commonalities between objects’ models places their representations in relation to other objects, thus leading to high generalization capabilities and lower storage demands.

However, although all neurophysiologic evidence suggests that in the human visual system quite a number of levels are realized, it has turned out that the design and/or learning of such deep hierarchical systems is a very difficult task. Most existing computer vision systems are 'flat' (e.g., having rather simple features - such as SIFT - as input and then applying some kind of SVM learning) and hence cannot make use of the advantages connected to deep hierarchies. Here in particular the generalization capabilities are crucial for any form of cognitive intelligence. As a consequence, we see the issue of establishing deep hierarchies as one major challenge for the establishment of truly cognitive systems.

The aim of the workshop is to bring together researchers from vision, robotics, machine learning, artificial intelligence, and neurophysiology to discuss existing obstacles in the design of deep hierarchies, possible solutions as well as perspectives for deep hierarchies in vision and robotics.

Research on reproducing kernel Hilbert spaces and kernel-based methods has witnessed a major impetus during the last two decades. Recent advances include kernels on structured data, powerful learning guarantees for kernel-based methods, and Hilbert-space embeddings of distributions. Some of the most lively NIPS and ICML workshops in recent years have dealt with applications where kernel approaches are popular, most notably multiple kernel learning, transfer learning, and multi-task learning. While kernel-based methods are well established in the machine learning practice, certain results in the underlying theory of RKHS remain relatively inaccessible to the ML community. Moreover, powerful tools for RKHS developed in other branches of mathematics, for instance in numerical analysis and probability, are less well known to machine learning researchers.

The proposed workshop represents an opportunity to bring together researchers in probability theory, mathematicians, and machine learning researchers working on RKHS methods. The goals of the workshop are threefold: first, to provide an accessible review and synthesis of classical results in RKHS theory from the point of view of functional analysis, probability theory, and numerical analysis. Second, to cover recent advances in RKHS theory relevant to machine learners (for instance, operator valued RKHS, kernels on time series, kernel embeddings of conditional probabilities). Third, to provide a forum for open problems, to elucidate misconceptions that sometimes occur in the literature, and to discuss technical challenges.

ICML 2012 Workshop on RKHS and kernel-based methods

ICML 2012 Workshop on RKHS and kernel-based methods

Workshop Topic:

The workshop is devoted to psychologically-motivated computational models of language acquisition. That is, models which are compatible with research in psycholinguistics, developmental psychology and linguistics.

Workshop History:

This is the sixth meeting of the Psychocomputational Models of Human Language Acquisition workshop following PsychoCompLA-2004, held in Geneva, Switzerland as part of the 20th International Conference on Computational Linguistics (COLING- 2004), PsychoCompLA-2005 as part of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL-2005) held in Ann Arbor, Michigan where the workshop shared a joint session with the Ninth Conference on Computational Natural Language Learning (CoNLL-2005), PsychoCompLA-2007 held in Nashville, Tennessee as part of the 29th meeting of the Cognitive Science Society (CogSci- 2007), PsychoCompLA-2008 held in Washington D.C., as part of the 30th meeting of the Cognitive Science Society (CogSci-2008), and PsychoCompLA-2009 held over two days before the 31st meeting of the Cognitive Science Society (CogSci-2009) in Amsterdam, Netherlands.

Workshop Description:

The workshop will present research and foster discussion centered around psychologically-motivated computational models of language acquisition, with an emphasis on the acquisition of syntax. In recent decades there has been a thriving research agenda that applies computational learning techniques to emerging natural language technologies and many meetings, conferences and workshops in which to present such research. However, there have been only a few (but growing number of) venues in which psychocomputational models of how humans acquire their native language(s) are the primary focus. Psychocomputational models of language acquisition are of particular interest in light of recent results in developmental psychology that suggest that very young infants are adept at detecting statistical patterns in an audible input stream. However, how children might plausibly apply statistical 'machinery' to the task of grammar acquisition, with or without an innate language component, remains an open and important question. One effective line of investigation is to computationally model the acquisition process and determine interrelationships between a model and linguistic or psycholinguistic theory, and/or correlations between a model's performance and data from linguistic environments that children are exposed to.

Topics and Goals:

Given the collocation of the workshop with the Input and Syntactic Acquisition workshop, submissions that present research related to the acquisition of syntax are strongly encouraged, though submissions on the computational modelling on any aspect of human language acquisition are welcome.

Specifically, submissions on (but not necessarily limited to) the following topics are welcome:

· Models that address the acquisition of word-order;

· Models that combine parsing and learning;

· Formal learning-theoretic and grammar induction models that incorporate psychologically plausible constraints;

· Comparative surveys that critique previously reported studies;

· Models that have a cross-linguistic or bilingual perspective;

· Models that address learning bias in terms of innate linguistic knowledge versus statistical regularity in the input;

· Models that employ language modeling techniques from corpus linguistics;

· Models that employ techniques from machine learning;

· Models of language change and its effect on language acquisition or vice versa;

· Models that employ statistical/probabilistic grammars;

· Computational models that can be used to evaluate existing linguistic or developmental theories (e.g., principles & parameters, optimality theory, construction grammar, etc.)

· Empirical models that make use of child-directed corpora such as CHILDES.

This workshop intends to bring together researchers from cognitive psychology, computational linguistics, other computer/mathematical sciences, linguistics and psycholinguistics working on all areas of language acquisition. Diversity and cross-fertilization of ideas is the central goal.

PsychoCompLA-2012

PsychoCompLA-2012

The workshop was funded by the Royal Society under the the Research Fellows International Scientific Seminars scheme, and the PASCAL2 network provided funding to cover the filming. The aim of the meeting was to bring together researchers from complex networks, and those working in machine learning and graph theory. The goal was to identify current challenges in complex networks analysis and identify possible methodologies for addressing them. The meeting was composed of four sessions:

1. methods for measuring and characterising complex network structure.
2. dynamic processes on complex networks.
3. complex network function prediction.
4. future directions, collaboration and networking.

The presentations were not intended to be lectures focused on specific research results. Instead they were expected to summarize state-of-the-art or accepted wisdom, challenge it and pose a provocative agenda for the discussions.

The field of computational biology has seen dramatic growth over the past few years. A wide range of high-throughput technologies developed in the last decade now enable us to measure parts of a biological system at various resolutions—at the genome, epigenome, transcriptome, and proteome levels. These technologies are now being used to collect data for an ever-increasingly diverse set of problems, ranging from classical problems such as predicting differentially regulated genes between time points and predicting subcellular localization of RNA and proteins, to models that explore complex mechanistic hypotheses bridging the gap between genetics and disease, population genetics and transcriptional regulation. Fully realizing the scientific and clinical potential of these data requires developing novel supervised and unsupervised learning methods that are scalable, can accommodate heterogeneity, are robust to systematic noise and confounding factors, and provide mechanistic insights.

The goals of this workshop are to i) present emerging problems and innovative machine learning techniques in computational biology, and ii) generate discussion on how to best model the intricacies of biological data and synthesize and interpret results in light of the current work in the field. We will invite several rising leaders from the biology/bioinformatics community who will present current research problems in computational biology and lead these discussions based on their own research and experiences. We will also have the usual rigorous screening of contributed talks on novel learning approaches in computational biology. We encourage contributions describing either progress on new bioinformatics problems or work on established problems using methods that are substantially different from established alternatives. Kernel methods, graphical models, feature selection, non-parametric models and other techniques applied to relevant bioinformatics problems would all be appropriate for the workshop. We are particularly keen on considering contributions related to the prediction of functions from genotypes and that target data generated from novel technologies such as gene editing and single cell genomics, though we will consider all submissions that highlight applications of machine learning into computational biology. The targeted audience are people with interest in learning and applications to relevant problems from the life sciences, including NIPS participants without any existing research link to computational biology.

Organizers

• Nicolo Fusi, Microsoft Research, Cambridge (USA)
• Anna Goldenberg, SickKids Research Institute program of Genetics and Genome Biology (Canada)
• Sara Mostafavi, University of British Columbia (Canada)
• Gerald Quon, MIT, Cambridge (USA)
• Oliver Stegle, EMBL (UK)

Program Committee:

• Alexis Battle, JHU
• Michael A. Beer, JHU
• Andreas Beyer, TU Dresden
• Karsten Borgwardt, ETH Zurich
• Gal Chechik, Gonda brain center, Bar Ilan University
• Chao Cheng, Dartmouth Medical School
• Manfred Claassen, ETH Zurich
• Florence d'Alche-Buc, Université d'Evry-Val d'Essonne, Genopole
• Saso Dzeroski, Jozef Stefan Institute
• Jason Ernst , UCLA
• Pierre Geurts, University of Liège
• James Hensman, The University of Sheffield
• Antti Honkela, University of Helsinki
• Laurent Jacob, Mines Paris Tech
• Samuel Kaski, Aalto University
• Seyoung Kim, CMU
• David Knowles, Stanford
• Anshul Kundaje, Stanford
• Neil Lawrence, University of Sheffield
• Su-In Lee, University of Washington
• Shen Li, Mount Sinai, New York
• Michal Linial, Hebrew University
• John Marioni, EMBL-EBI
• Martin Renqiang Min, NEC Labs America
• Yves Moreau, KU Leuven
• Alan Moses, University of Toronto
• Bernard Ng, UBC
• William Noble, University of Washington
• Uwe Ohler, MDC Berlin & Humboldt University
• Yongjin Park, MIT
• Leopold Parts, University of Toronto
• Dana Pe'er, Columbia University
• Nico Pfeifer, Max Planck Institute
• Magnus Rattray, University of Manchester
• Simon Rogers, University of Glasgow
• Juho Rousu, Aalto University
• Guido Sanguinetti, University of Edinburgh
• Alexander Schliep, Rutgers University
• Jean-Philippe Vert, Ecole des Mines de Paris
• Jinbo Xu, Toyota Technological Institute of Chicago
• Chun (Jimmie) Ye , UCSF

Solving optimization problems with ultimately discretely solutions is becoming increasingly important in machine learning: At the core of statistical machine learning is to infer conclusions from data, and when the variables underlying the data are discrete, both the tasks of inferring the model from data, as well as performing predictions using the estimated model are discrete optimization problems. Many of the resulting optimization problems are NP-hard, and typically, as the problem size increases, standard off-the-shelf optimization procedures become intractable.

Fortunately, most discrete optimization problems that arise in machine learning have specific structure, which can be leveraged in order to develop tractable exact or approximate optimization procedures. For example, consider the case of a discrete graphical model over a set of random variables. For the task of prediction, a key structural object is the "marginal polytope," a convex bounded set characterized by the underlying graph of the graphical model. Properties of this polytope, as well as its approximations, have been successfully used to develop efficient algorithms for inference. For the task of model selection, a key structural object is the discrete graph itself. Another problem structure is sparsity: While estimating a high-dimensional model for regression from a limited amount of data is typically an ill-posed problem, it becomes solvable if it is known that many of the coefficients are zero. Another problem structure, submodularity, a discrete analog of convexity, has been shown to arise in many machine learning problems, including structure learning of probabilistic models, variable selection and clustering. One of the primary goals of this workshop is to investigate how to leverage such structures.

The focus of this yearâÂÂs workshop is on the interplay between discrete optimization and machine learning: How can we solve inference problems arising in machine learning using discrete optimization? How can one solve discrete optimization problems involving parameters that themselves are estimated from training data? How can we solve challenging sequential and adaptive discrete optimization problems where we have the opportunity to incorporate feedback (online and active learning with combinatorial decision spaces)? We will also explore applications of such approaches in computer vision, NLP, information retrieval etc.
Organizers

• Andreas Krause, ETH Zurich
• Pradeep Ravikumar, University of Texas, Austin
• Jeff A. Bilmes, University of Washington
• Stefanie Jegelka, Max Planck Institute for Intelligent Systems, Tuebingen

Bayesian nonparametric methods are an expanding part of the machine learning landscape. Proponents of Bayesian nonparametrics claim that these methods enable one to construct models that can scale their complexity with data, while representing uncertainty in both the parameters and the structure. Detractors point out that the characteristics of the models are often not well understood and that inference can be unwieldy. Relative to the statistics community, machine learning practitioners of Bayesian nonparametrics frequently do not leverage the representation of uncertainty that is inherent in the Bayesian framework. Neither do they perform the kind of analysis --- both empirical and theoretical --- to set skeptics at ease. In this workshop we hope to bring a wide group together to constructively discuss and address these goals and shortcomings.

Organizers

• David B. Dunson, Duke University
• Ryan Prescott Adams, Harvard University
• Emily B. Fox, University of Pennsylvania

Advisory Panel

• David B. Dunson, Duke University
• Zoubin Ghahramani, University of Cambridge
• Michael I. Jordan, University of California at Berkeley
• Peter Orbanz, Cambridge University
• Yee Whye Teh, University College London
• Larry Wasserman, Carnegie Mellon University

From high-throughput biology and astronomy to voice analysis and medical diagnosis, a wide variety of complex domains are inherently continuous and high dimensional.
The statistical framework of copulas offers a flexible tool for modeling highly non-linear multivariate distributions for continuous data. Copulas are a theoretically and
practically important tool from statistics that explicitly allow one to separate the dependency structure between random variables from their marginal distributions.
Although bivariate copulas are a widely used tool in finance, and have even been famously accused of "bringing the world financial system to its knees" (Wired Magazine, 2009), the use of copulas for high dimensional data is in its infancy.

While studied in statistics for many years, copulas have only recently been noticed by a number of machine learning researchers, with this "new" tool appearing in the recent leading machine learning conferences (ICML, UAI and NIPS). The goal of this workshop is to promote the further understanding and development of copulas for the kinds of complex modeling tasks that are the focus of machine learning. Specifically, the goals of the workshop are to:

• draw the attention of machine learning researchers to the important framework of copulas
• provide a theoretical and practical introduction to copulas
• identify promising research problems in machine learning that could exploit copulas
• bring together researchers from the statistics and learning communities working in this area.

The target audience includes leading researchers from academia and industry, with the aim of facilitating cross fertilization between different perspectives.

Organizers

• Gal Elidan, The Hebrew University of Jerusalem
• Zoubin Ghahramani Cambridge University and Carnegie Mellon University
• John Lafferty, University of Chicago and Carnegie Mellon University