Workshop Description and Main Topics

Research in the last few years has focused on approximate computing as a means to overcome the energy scaling barriers of computer systems and the worsened reliability issues of the underlying hardware. The new paradigm attempts to utilize the inherent error resilience of key algorithms in many application domains such as signal processing, multimedia, data analytics, deep learning, hpc simulations and computational engineering, among others for relaxing the conservative high numerical accuracy or strict reliability constraints and allowing the use of less power hungry architectures and adequately reliable/precise computing schemes. For instance, fully accurate arithmetic in specific phases of a computation in those applications may have only a marginal effect on output quality, especially if combined with error correction frameworks such as iterative refinement. Thus, accurate execution may be traded off with lower energy consumption by providing the ability to scale supply voltage below nominal values or to use lower precision arithmetic (i.e. 8 or 16 bit).

Rather than focusing on a single layer, designing such systems in a general-purpose computing environment requires a holistic view of all layers from algorithms, programming models, system software, and hardware down to the transistor level. This workshop is an inter-disciplinary effort to bring together researchers from the areas of mathematics, computer science, computer and electrical engineering to discuss challenges, risks and opportunities of approximate and transprecision computing in all design layers. Papers will be published online in the WAPCO website, but not in proceedings. Submitting to WAPCO will not preclude future publication opportunities. Topics include but are not limited to the following:

  • Formal and mathematical methods for approximate computing
  • Programming languages and models for approximate computing
  • Compiler and system software support for approximate computing
  • Hardware, processor and memory techniques for supporting approximate computing
  • Hardware-software interaction for approximate computing
  • Applications that can benefit from approximate computing
  • Simulation and modeling techniques for approximate computing
  • Experimental results, models and error resilient methods for operation beyond pessimistic operating points
  • Position papers on the potential and limitations of approximate computing

Format of the Workshop

We are planning for a full-day workshop, which will include peer-reviewed papers, invited talks and a panel. We are planning for 8 to 10 talks, which we expect to be attended by 50+ people. There will be an open call for papers, and a peer-to-peer review process.

General Chair

Dimitrios Nikolopoulos Queen's University Belfast, UK
George Karakonstantis Queen's University Belfast, UK

Program Chairs

Cristiano Malossi IBM Research - Zurich, Switzerland
Nikolaos Bellas University of Thessaly, Greece
Frank K. Gurkaynak ETH Zurich, Switzerland

Web and Submission Chair

Charles Gillan Queen's University Belfast, UK

Program Committee

Hans Vandierendonck Queen's University Belfast, UK
Andreas BurgEPFL, Switzerland
Uwe NaumannRWTH-Aachen, Germany
Christos D. AntonopoulosUniversity of Thessaly, Greece
Spyros LalisUniversity of Thessaly, Greece
Holger Froening University of Heidelberg, Germany
Vincent HeuvelineUniversity of Heidelberg, Germany
Enrique Quintana OrtiUniversitat Jaume I de Castellon, Spain
Pedro TrancosoUniversity of Cyprus
Stefano Di Carlo Politecnico di Torino
Ramon Canal UPC, Barcelona
Sek Chai SRI, International, USA
Lukas Sekanina Brno University of Technology, Czech Rep.
Alberto Bosio LIRMM/CNRS, France