- From: Adolfo UPV <admarus@dsic.upv.es>
- Date: Thu, 02 Apr 2015 18:22:54 +0200
- To: Adolfo UPV <admarus@dsic.upv.es>
[Apologies for cross-postings] ################################################################## LMCE 2015 # Second International Workshop on Learning over Multiple Contexts @ ECML 2015 The value of model reuse A workshop held in conjunction with the ECML PKDD 2015, Porto, Portugal, 7-11 September 2015 http://www.dsic.upv.es/~flip/LMCE2015/ ################################################################## It is our pleasure to present this 2nd LMCE workshop, following up on the first workshop (http://users.dsic.upv.es/~flip/LMCE2014/) success with around 40 participants, 18 submissions, 13 regular papers and 3 work-in-progress contributions. As new this year, - This second edition will focus on reusing models. Indeed when a change of context is observed during deployment, it is often hard to train a new model, while there exist models trained in different conditions. - A challenge competition on real data (bike sharing) will be jointly organised with the workshop in order to give even more importance to model reuse over multiple contexts. We expect this framework will foster understanding, comparisons and discussions (see http://reframe-d2k.org/Challenge for further details on this competition). === Call for Papers === Adaptive reuse of learnt knowledge is of critical importance in the majority of knowledge-intensive application areas, particularly when the context in which the learnt model operates can be expected to vary from training to deployment. In machine learning this has been studied, for example, in relation to variations in class and cost skew in (binary) classification, leading to the development of tools such as ROC analysis to adjust decision thresholds to operating conditions concerning class and cost skew. More recently, considerable effort has been devoted to research on transfer learning, domain adaptation, and related approaches. Given that the main business of predictive machine learning is to generalise from training to deployment, there is clearly scope for developing a general notion of operating context. Without such a notion, a model predicting sales in Prague for this week may perform poorly in Nancy for next Wednesday. The operating context has changed in terms of location as well as resolution. While a given predictive model may be sufficient and highly specialised for one particular operating context, it may not perform well in other contexts. If sufficient training data for the new context is available it might be feasible to retrain a new model; however, this is generally not a good use of resources, and one would expect it to be more cost-effective to learn one general, versatile model that effectively generalizes over multiple and possibly previously unseen contexts. The aim of this workshop is to bring together people working in areas related to versatile models and model reuse over multiple contexts. Given the advances made in recent years on specific approaches such as transfer learning, an attempt to start developing an overarching theory is now feasible and timely, and can be expected to generate considerable interest from the machine learning community. Papers are solicited in all areas relating to model reuse and model generalisation including the following areas: * Context-aware applications and recommender systems. * Cost-sensitive Learning. * Dataset shift, including concept drift and covariate shift. * Domain adaptation. * Employing background knowledge. * Formats and tools for model exchange, transformation and reuse, such as PMML. * Learning with different feature granularities or dimensions, quantification. * Logical approaches to model reuse: abduction, theory revision, ILP, . * Meta-Learning. * Multi-task learning, co-learning. * ROC analysis. * Soft classifiers and Probability Estimators. * Transductive Learning. * Transfer Learning. * Volatile information sources, adversarial learning. === Submission of Papers === We welcome submissions describing work in progress as well as more mature work related to learning over multiple contexts and model reuse. Submissions should be between 6 and 16 pages in the same format as the ECML-PKDD conference (LNAI). Authors of accepted papers will be asked to prepare a poster, and selected authors will be given the opportunity of a plenary presentation during the workshop. Submission website: https://www.easychair.org/conferences/?conf=lmce2015 Papers will be selected by the program committee according to the quality of the submission and its relevance to the workshop topics. All accepted papers will be published on the workshop web site. The publication of a selected set of papers for a special volume or a journal issue is considered, but this will depend on the success and overall results of the workshop. === Important Dates === Submission (workshop papers): June 8, 2015 Notification of acceptance: July 6, 2015 Camera Ready copies: August 3, 2015 Workshop/Challenge dates: September 11, 2015 === Program Committee === Chowdhury Farhan Ahmed, University of Strasbourg, France Wouter Duivesteijn, Technische Universitat Dortmund, Germany Cesar Ferri, Technical University of Valencia, Spain Amaury Habrard, University of Saint-Etienne, France José Hernandez-Orallo, Technical University of Valencia, Spain Francisco Herrera, University of Granada, Spain Sinno Jialin, Nanyang Technological University, Singapore Antonio M. Lopez, Universitat Autonoma de Barcelona, Spain Dragos Margineantu, Boeing Research and Technology, U.S.A. Weike Pan, Shenzhen University, China. Huimin Zhao, University of Wisconsin-Milwaukee, U.S.A. === Organising Committee === Nicolas Lachiche, University of Strasbourg, France Meelis Kull, University of Bristol, UK Adolfo Martínez-Usó, Universitat Politècnica de Valencia, Spain For more information visit http://www.dsic.upv.es/~flip/LMCE2015/
Received on Wednesday, 8 April 2015 12:23:59 UTC