The invited speakers are Avrim Blum,
Gábor Lugosi and
(1) ALT 2016 Keynote Speaker Avrim Blum: Learning about Agents and Mechanisms from Opaque Transactions.
In this talk I will discuss some learning problems coming from the area of algorithmic economics. I will focus in particular on settings known as combinatorial auctions in which agents have preferences over items or sets of items, and interact with an auction or allocation mechanism that determines what items are given to which agents. We consider the perspective of an outside observer who each day can only see which agents show up and what they get, or perhaps just which agents' needs are satisfied and which are not. Our goal will be from observing a series of such interactions to try to learn the agent preferences and perhaps also the rules of the allocation mechanism.
As an example, consider observing web pages where the agents are advertisers and the winners are those whose ads show up on the given page. Or consider observing the input-output behavior of a cloud computing service, where the input consists of a set of agents requesting service, and the output is a partition of them into some whose requests are actually fulfilled and the rest that are not - due to overlap of their resource needs with higher-priority requests. From such input-output behavior, we would like to learn the underlying structure. We also consider a classic Myerson single-item auction setting, where from observing who wins and also being able to participate ourselves we would like to learn the agents' valuation distributions.
In examining these problems we will see connections to decision-list learning and to Kaplan-Meier estimators from medical statistics.
This talk is based on work joint with Yishay Mansour and Jamie Morgenstern.
Avrim Blum is Professor of Computer Science at Carnegie Mellon University. His main research interests are in Foundations of Machine Learning and Data Mining, Algorithmic Game Theory (including auctions, pricing, dynamics, and connections to machine learning), the analysis of heuristics for computationally hard problems, and Database Privacy. He has served as Program Chair for the IEEE Symposium on Foundations of Computer Science (FOCS) and the Conference on Learning Theory (COLT). He was recipient of the Sloan Fellowship, the NSF National Young Investigator Award, the ICML/COLT 10-year best paper award, and the Herbert Simon Teaching Award, and is a Fellow of the ACM.
(2) DS 2016 Keynote Speaker Kristian Kersting: Collective attention on the web.
It's one of the most popular YouTube videos ever produced, having been viewed more than 840 million times. It's hard to understand why this clip is so famous and actually went viral, since nothing much happens. Two little boys, Charlie and Harry, are sitting in a chair when Charlie, the younger brother, mischievously bites Harry's finger. There's a shriek and then a laugh. The clip is called "Charlie Bit My Finger--Again!"
Generally, understanding the dynamics of collective attention is central to an information age where millions of people leave digital footprints everyday. So, can we capture the dynamics of collective attention mathematically? Can we even gain insights into the underlying physical respectively social processes? Is it for instance fair to call the video "viral" in an epidemiological sense?
In this talk I shall argue that computational methods of collective attention are not insurmountable. I shall review the methods we have developed to characterize, analyze, and even predict the dynamics of collective attention among millions of users to and within social media services. For instance, we found that collective attention to memes and social media grows and subsides in a highly regular manner, well explained by economic diffusion models. Using mathematical epidemiology, we find that so-called "viral" videos show very high infection rates and, hence, should indeed be called viral. Moreover, the spreading processes may also be related to the underlying network structures, suggesting for instance a physically plausible model of the distance distributions of undirected networks. All this favors machine learning and discovery science approaches that produce physically plausible models.
This work was partly supported by the Fraunhofer ICON project SoFWIReD and by the DFG Collaborative Research Center SFB 876 project A6. It is based on joint works with Christian Bauckhage, Fabian Hadiji and Rastegarpanah.
Kristian Kersting is an Associate Professor in the Computer Science Department at the Technical University of Dortmund. His main research interests are data mining, machine learning, and statistical relational artificial intelligence, with applications to medicine, plant phenotpying, traffic, and collective attention. He gave several tutorials at top conferences and co-chaired BUDA, CMPL, CoLISD, MLG, and SRL as well as the AAAI Student Abstract track and the Starting AI Research Symposium (STAIRS). Together with Stuart Russell (Berkeley), Leslie Kaelbling (MIT), Alon Halevy (Goolge), Sriraam Natarajan (Indiana) and Lilyana Mihalkova (Google) he cofounded the international workshop series on Statistical Relational AI. He served as area chair/senior PC for several top conference and co-chaired ECML PKDD 2013, the premier European venue for Machine Learning and Data Mining. Currently, he is an action editor of JAIR, AIJ, DAMI, and MLJ as well as on the editorial board of NGC.
(3) Joint ALT 2016 and DS 2016 Keynote Speaker John Shawe-Taylor: Margin based structured output learning.
Structured output learning has been developed to borrow strength across multidimensional classifications. There have been approaches to bounding the performance of these classifiers based on different measures such as microlabel errors with a fixed simple output structure. We present a different approach and analysis starting from the assumption that there is a margin attainable in some unknown or fully connected output structure. The analysis and algorithms flow from this assumption but in a way that the associated inference becomes tractable while the bounds match those attained were we to use the full structure. There are two variants depending on how the margin is estimated. Experimental results show the relative strengths of these variants, both algorithmically and statistically.
John Shawe-Taylor is a professor at the University College London where he directs the Centre for Computational Statistics and Machine Learning and heads the Department of Computer Science. His research has contributed to a number of fields ranging from graph theory through cryptography to statistical learning theory and its applications. However, his main contributions have been in the development of the analysis and subsequent algorithmic definition of principled machine learning algorithms founded in statistical learning theory. He has co-authored two influential text books on kernel methods and support vector machines. He has also been instrumental in coordinating a series of influential European Networks of Excellence culminating in the PASCAL networks.
(4) ALT 2016 Tutorial Speaker Gábor Lugosi: How to Estimate the Mean of a Random Variable?
Given n independent, identically distributed copies of a random variable, one is interested in estimating the expected value. Perhaps surprisingly, there are still open questions concerning this very basic problem in statistics. In this talk we are primarily interested in non-asymptotic sub-Gaussian estimates for potentially heavy-tailed random variables. We discuss various estimates and extensions to high dimensions. We apply the estimates for statistical learning and regression function estimation problems. The methods improve on classical empirical minimization techniques.
This talk is based on joint work with Emilien Joly, Luc Devroye, Matthieu Lerasle, Roberto Imbuzeiro Oliveira, and Shahar Mendelson.
Gábor Lugosi is an ICREA research professor at the Department of Economics, Pompeu Fabra University, Barcelona. His research main interests include the theory of machine learning, combinatorial statistics, inequalities in probability, random graphs and random structures, and information theory. He has co-authored monographs on pattern classification, on online learning, and on concentration inequalities.
(5) DS 2016 Tutorial Speaker Gianluca Bontempi: Perspectives of feature selection in bioinformatics: from relevance to causal inference.
A major goal of the scientific activity is to model real phenomena by studying the dependency between entities, objects or more in general variables. Sometimes the goal of the modeling activity is simply predicting future behaviors. Sometimes the goal is to understand the causes of a phenomenon (e.g. a disease). Finding causes from data is particular challenging in data mining and bioinformatics where often the number of features (e.g. number of SNPs or microarray probes) is huge with respect to the number of samples. In this context, even when experimental interventions are possible, performing thousands of experiments to discover causal relationships between thousands of variables is not practical. Dimensionality reduction techniques have been largely discussed and used in bioinformatics to deal with the curse of dimensionality. However, most of the time these techniques focus on improving prediction accuracy, neglecting causal aspects. This talk will discuss a major open issue : may feature selection techniques be useful also for causal feature selection? Is prediction accuracy compatible with causal discovery? Some results based on an information theory approach will be used to illustrate the issue. Gianluca Bontempi is Full Professor in the Computer Science Department at the Université Libre de Bruxelles (ULB), Brussels, Belgium, co-head of the ULB Machine Learning Group and Director of (IB)2, the ULB/VUB Interuniversity Institute of Bioinformatics in Brussels. His main research interests are big data mining, machine learning, bioinformatics, causal inference, predictive modelling and their application to complex tasks in engineering (forecasting, fraud detection) and life science. He was Marie Curie fellow researcher, he was awarded in two international data analysis competitions and he took part to many research projects in collaboration with universities and private companies all over Europe. He is author of more than 200 scientific publications, associate editor of PLOS One, member of the scientific advisory board of Chist-ERA and IEEE Senior Member. He is also co-author of several open-source software packages for bioinformatics, data mining and prediction.