Second International Workshop on Energy Efficient Scalable Data Mining and Machine Learning
Co-located with ECML PKDD 2019
September 16, 2019 - Würzburg, Germany
Room 1.002 (Hubland campus of the University of Würzburg)
KEYNOTE SPEAKERS
TU Dortmund University
Arm ML Research Lab and Harvard University
Ekkono Solutions
This workshop aims to bring together people from different areas and backgrounds in data mining and machine learning that have a common interest in energy efficiency, scalability, and edge computing.
For the past years, the main concern in machine learning had been to create highly accurate models, without considering the high computational requirements involved. This has lead to a scenario where most of the machine learning prediction is done in the cloud, incurring in security concerns and increased latency.
However, there is an increasing trend in machine learning which focuses on building models that are able to run in the edge. For instance, Google has released the first speech recognition model running directly on the device, improving latency and reducing energy consumed by the network connectivity. Another example is TensorLite, a powerful tool to deploy models on mobile and IoT devices.
The goal with this workshop is to promote green machine learning even further, by creating a half-day workshop where researchers in different machine learning and data mining areas can bring together their ideas, present them in front of a heterogeneous crowd, and have interesting debates on how to advance machine learning into a more scalable future.
We accept original work, already completed, or in progress. Position papers and extended abstracts are also considered.
Accepted papers (except for extended abstracts) will be published in ECML-PKDD 2019 Workshop proceedings.
For the past years, the main concern in machine learning had been to create highly accurate models, without considering the high computational requirements involved. This has lead to a scenario where most of the machine learning prediction is done in the cloud, incurring in security concerns and increased latency.
However, there is an increasing trend in machine learning which focuses on building models that are able to run in the edge. For instance, Google has released the first speech recognition model running directly on the device, improving latency and reducing energy consumed by the network connectivity. Another example is TensorLite, a powerful tool to deploy models on mobile and IoT devices.
The goal with this workshop is to promote green machine learning even further, by creating a half-day workshop where researchers in different machine learning and data mining areas can bring together their ideas, present them in front of a heterogeneous crowd, and have interesting debates on how to advance machine learning into a more scalable future.
We accept original work, already completed, or in progress. Position papers and extended abstracts are also considered.
Accepted papers (except for extended abstracts) will be published in ECML-PKDD 2019 Workshop proceedings.
Key Dates
- Workshop paper submission deadline: Monday, July 1, 2019
- Workshop paper acceptance notification: Friday, July 19, 2019
- Workshop paper camera-ready deadline: Monday, July 26, 2019
- Workshop date: Friday, September 16, 2019
Workshop Chairs
- Eva García-Martín, Blekinge Institute of Technology
- Albert Bifet, Telecom-ParisTech
- Crefeda Faviola Rodrigues, University of Manchester
- Heitor Murilo Gomes, Telecom-ParisTech
Steering Committee
- Ricardo Baeza-Yates, NTENT
- Christian Nordahl, Blekinge Institute of Technology
- Veselka Boeva, Blekinge Institute of Technology
- Elena Tsiporkova, Sirris (Collective Center for the Belgian technological industry)
- Niklas Lavesson, Jönköping University
- Håkan Grahn, Blekinge Institute of Technology
- Emiliano Casalicchio, Sapienza University of Rome