***********************************************************************
                           Call for Papers
                   2nd International Workshop on 
    Active Learning: Applications, Foundations and Emerging Trends
                http://www.uni-kassel.de/go/al-ijcnn

                         co-located with the 
     International Joint Conference on Neural Networks (IJCNN 2017) 
              Anchorage, Alaska, USA, May 14–19, 2017
***********************************************************************

Topic
=====
Science, technology, and commerce increasingly recognize the
importance of machine learning approaches for data-intensive, evidence
based decision making. While the number of machine learning
applications and the volume of data increases, resources like the
capacities of processing systems or human supervisors remain limited.
This makes active learning techniques an important but challenging
research topic. Active Learning bridges the gap between data-centric
and user-centric approaches by optimizing their interaction, e.g., by
selecting the most  relevant information, by performing the most
informative experiment, or by selecting solely the most informative
data for processing. Thereby, it enables efficient allocation of
limited resources, thus reducing costs in terms of time (e.g., human
effort or processing time) and money.

Active Learning is a very useful methodology in on-line industrial 
applications to minimize the effort for sample annotation and 
measurements of "target" values (e.g., quality criteria). It further 
reduces the computation load of machine learning and data mining
tools, as embedded models are only updated based on a subset of samples
selected by the implemented active learning technique. Especially, in 
cost-intensive areas like medical applications (e.g., diagnostic
support, brain-computer interfaces) the efficient use of expert
knowledge is crucial.

However, there are several recent research directions, open problems,
and challenges in active learning, which ideally should be addressed
and discussed in this workshop.

Thus, we welcome contributions on active learning that address aspects
including, but not limited to:
 -  new active learning methods and models,
 -  active learning for recent complex model structures, such as (deep)
neural networks or extreme learning machines,
 -  applications and real-world deployment of active learning, new 
interactive learning protocols and application scenarios, e.g., 
brain-computer interfaces, crowdsourcing, etc.,
 -  evaluation of active learning and comparative studies,
 -  active learning for big data and evolving datastreams,
 -  active learning applications, e.g., in industry,
 -  active class or feature selection,
 -  active filtering, forgetting, or resampling,
 -  active, user-centric approaches for selection of information,
 -  combinations with change detection or transfer learning, or
 -  innovative use of active learning techniques, e.g., for detection
of outliers, frauds, or attacks.

Website
=======
http://www.uni-kassel.de/go/al-ijcnn

Important Dates
===============
Submission Deadline:  Feb 2, 2017
Notification Due:     Mar 9, 2017
Final Version Due:    Apr 10, 2017
Conference (IJCNN):   May 14, 2017 - May 19, 2017

Paper Submission
================
Up to 8 pages, open access through ceur-ws.org (google scholar indexed)
https://easychair.org/conferences/?conf=alijcnn17

Organizers
==========
 - Georg Krempl (University Magdeburg, Germany)
 - Vincent Lemaire (Orange Labs, France)
 - Robi Polikar (Rowan University, USA)
 - Bernhard Sick (University of Kassel, Germany)
 - Daniel Kottke (University of Kassel, Germany)
 - Adrian Calma (University of Kassel, Germany)

Liebe Grüße
Daniel

--
Mailing-Liste: wi@lists.kit.edu
Administrator: wi-request@lists.kit.edu
Konfiguration: https://www.lists.kit.edu/wws/info/wi

Betreff:	[WI] CfP: 2nd Int. Workshop on Active Learning: Applications, Foundations and Emerging Trends @IJCNN2017
Datum:	Wed, 11 Jan 2017 17:57:39 +0100
Von:	Daniel Kottke <daniel.kottke@uni-kassel.de>
Antwort an:	Daniel Kottke <daniel.kottke@uni-kassel.de>
An:	kdml@cs.uni-kassel.de, ak-kd-list@lists.uni-karlsruhe.de, fg-db@informatik.uni-rostock.de, wi@aifb.uni-karlsruhe.de