Second Workshop on Energy-Efficient Machine Learning (E2ML)

Scope of the Workshop

Advances in the machine learning (ML) and its deployment in a wide range of systems for various applications. This has stirred interest in the design of various devices ranging from cloud servers to miniature IoT devices equipped with smart capabilities with embedded ML. One of the major hurdles to be addressed for an efficient design of such systems is to managed the limited available resources. The second edition of Workshop on Energy-Efficient Machine Learning (E2ML) workshop focuses on the design strategies to minimize the footprint and efficient management of resources through advanced computing techniques as well as resource management. The topics of interest to this workshop are:

  • In-memory computing
  • Neuromorphic computing
  • Approximate computing for ML applications
  • Power management for ML architectures
  • Emerging memory technologies and its applicability in ML applications
  • Spiking neural networks
  • Learning algorithms on embedded systems
  • Hardware-software cross-layer co-design
  • Distributed ML algorithms and hardware for real-time performance
  • Any other relevant topic related to design of hardware for ML and optimizing ML for resource constrained systems is within the scope of the workshop. Papers that showcases interesting analysis regarding embedded ML and the possible ways to extend are also welcome.

    Program Schedule

    The workshop will be carried out on a virtual platform platform and the details for access to virtual sessions will be announced to workshop (IGSC 2020 conference) registrants shortly before Oct 19, 2020.

    Workshop Agenda - Monday, October 19 (All times are EST)

  • 8:00 am - 8:20 am "Profiling Energy Consumption of Deep Neural Networks on NVIDIA Jetson Nano" by Stephan Holly, Alexander Wendt and Martin Lechner (Vienna University of Technology, Austria).
  • 8:20 am - 8:40 am "A Comprehensive Review of ML-based Time-Series and Signal Processing Techniques and their Hardware Implementations" by Abhijitt Dhavlle and Sai Manoj P D (George Mason University, USA).
  • 8:40 am - 9:00 am "Intermittent Inference with Nonuniformly Compressed Multi-Exit Neural Network for Energy Harvesting Powered Devices", by Jintong Hu (University of Pittsburgh, USA).
  • 9:00 am - 9:20 am "Investigating IDS Usage to Prevent DDoS Attack in a Cloud Computing Environment", by Muhammad Ubaid Ur Rehman and Qamar Mahmood (Capital University of Science and Technology, Pakistan).
  • 9:20 am - 9:40 am "Machine Learning Based Multi-Objective Design Space Exploration for Manycore Systems", by Ryan Kim (Colarado State University).
  • Submission Guidelines

    All papers must be original and not simultaneously submitted to another journal or conference. Full papers should be of length up to 6 single-spaced double-column pages using 10-point size font on 8.5x11 inch pages (IEEE conference style) including figures, tables, and references.

  • IEEE templates can be found here.
  • Full paper submission deadline: September 1st, 2020 Closed
  • Notification of acceptance: September 15th, 2020
  • Final manuscript due: September 21st, 2020
  • Paper has to be submitted over easychair.


    The workshop proceedings will be published in IEEE Explore.

    Due to COVID-19 related delays in finalizing the conference program:

  • We will temporarily make a PDF copy of the Workshop Proceedings (without copyright form collection) available on the IGSC website during the conference so that the conference registrants can have access to the papers.
  • Once the IEEE copyright forms are ready, we will be in touch to collect the final manuscripts and signed copyright forms in order to have the IGSC Workshop Proceedings available on IEEE Xplore.
  • Venue

    Given these unprecedent circumstances, the IGSC and E2ML workshop will happen virtually.


    All questions about submissions should be directed to Sai Manoj P D

    Workshop Organizer

    Sai Manoj P D, George Mason University