AxC21: 6th Workshop on Approximate Computing

Virtual Event: November 04th, 2021

Approximate Computing leverages the intrinsic error resilience of applications to inaccuracy in their inner calculations, in order to achieve a required trade-off between efficiency, in terms of performance and power demanding, and acceptable error of returned results. In particular, for audio, image and video processing, data mining and information retrieval, approximate results turn out hard to distinguish from perfect ones. In recent years, Approximate Computing applicability is broadening and it has been representing a breakthrough in many scientific areas. Suitable solutions come from approximate arithmetic operators, implemented both at hardware and software level, but from unreliable memory architectures, integrated circuit test, compilers and many others.

This year event will be in conjunction with ICCAD'21

You are invited to participate and submit your contributions to the AxC Workshop. The workshop's areas of interest include (but are not limited to) the following topics:

  • Approximation for Deep Learning Applications
  • Approximation techniques for emerging processor and memory technologies
  • Approximation-induced error modeling and propagation
  • Approximation in Edge computing applications
  • Approximation in HPC and Embedded systems
  • Architectural support for Approximation
  • Dependability of approximate circuits and systems
  • Design automation of Approximate architectures
  • Design of reconfigurable Approximate
  • Test and fault tolerance of approximate systems
  • Hardware/software co-design of Approximate systems
  • Language, compiler, and operating system support for approximate architectures
  • Modeling, specification, and verification of approximate circuits and systems
  • Safety and reliability applications of approximate computing
  • Security in the context of Approximation
  • Software-based fault tolerant technique for approximate computing
  • Techniques for monitoring and controlling approximation quality
  • Approximation in Near-Memory and Database Processing

    • Supporter

    This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 956090.

    Technical Program

    UTC-7 CET UTC+8
    07:20–07:30 15:20–15:30 22:20–22:30

    Opening (Welcome Message from Chairs)
    07:30–09:50 15:30–17:50 22:30–00:50

    Invited Talks

    (40 minutes each (with Q&A) + 10 minute break)
    1. Approximate Computing for Low Power Circuits and Systems
      Weiqiang Liu - Nanjing University of Aeronautics and Astronautics, Nanjing, China
    2. Automated and Sound Approximation of Numerical Kernels
      Eva Darulova - Max Planck Institute for Software Systems (MPI-SWS)
    3. Approximate Computing – The Holy Grail of Low-Energy Electronic Systems?
      Jari Nurmi - Tampere University
    09:50–10:00 17:50–18:00 00:50–01:00 Break
    10:00–11:00 18:00–19:00 01:00–02:00

    Session 1: Approximate computing in Machine Learning and DSP Applications

    (15m + 5m Q&A each)
    1. Neural Network Hardware Acceleration Based on Hybrid Approximate Multipliers
      Authors: Siyuan Liang, Ke Chen, Pengfei Huang, Peipei Yin and Weiqiang Liu
    2. CNN Compression through Retraining-free Weight Sharing
      Authors: Etienne Dupuis, David Novo, Ian O'Connor and Alberto Bosio
    3. Combining Logic Refactoring and Approximate Computing to Power- and Area-Efficient Gaussian Filter Designs
      Authors: Marcio Monteiro, Ismael Seidel, Mateus Grellert, Jose Luis Güntzel, Leonardo Soares and Cristina Meinhardt
    11:00–11:10 19:00–19:10 02:00–02:10 Break
    11:10–12:10 19:10–20:10 02:10–03:10

    Session 2: Approximate Design Techniques

    (15m + 5m Q&A each)
    1. Approximate PIM: An IMPLY-based Approximate Adder
      Authors: Seyed Erfan Fatemieh, Mohammad Reza Reshadinezhad and Nima Taherinejad
    2. Assertion-aware based approximate computing
      Authors: Moreno Bragaglio, Samuele Mori, Samuele Germiniani, Alberto Bosio, Marcello Traiola and Graziano Pravadelli
    3. Task Scheduling on Mixed-Precision MPSoC
      Authors: Ahmad Sadigh Baroughi, Stefan Huemer, Hadi Shahriar Shahhoseini and Nima Taherinejad
    12:10–12:20 20:10–20:20 03:10–03:20 Break
    12:20–13:20 20:20–21:20 03:20–04:20

    Session 3: Approximate Synthesis for FPGA targets

    (15m + 5m Q&A each)
    1. Approximation Space Exploration for FPGA Accelerators Using Fault Injection
      Authors: Ioannis Tsounis, Athanasios Papadimitriou and Mihalis Psarakis
    2. Exploring a Decision Tree Synthesis Flow for Approximate Circuits
      Authors: Isac Souza Campos, Augusto Berndt, Brunno Abreu, Jonata Carvalho, Mateus Grellert and Cristina Meinhard
    3. A Catalog-based AIG-Rewriting Approach to the Design of Approximate Components
      Authors: Mario Barbareschi, Salvatore Barone, Nicola Mazzocca and Alberto Moriconi
    13:20–13:30 21:20–21:30 04:20–04:30

    Closing Session

    Important Dates

    • Submission deadline : September 15th, 2021 EXTENDED to October 22nd, 2021
    • Notification of acceptance : October 08th, 2021 October 28th, 2021

    Paper Submission

    The Workshop accepts Extended Abstract submissions (of at most two pages). Each submission has to be in a standard IEEE format. The IEEE template can be found here.

    Submit your paper on EasyChair

    Organizing Committee

    General Chairs: Alexandra Kourfali - Stuttgart U (DE)
    Alberto Bosio - ECL (FR)
    Program Chairs: Alessandro Savino - POLITO (IT)
    Jürgen Teich - FAU (DE)
    Review Chair: Mario Barbareschi - UniNa (IT)

    Steering Committee

    Jie HanU. Alberta (CA)
    Sybille HellebrandPaderborn U (DE)
    Jörg HenkelKIT (DE)
    Anand RaghunathanPurdue U (USA)
    Kaushik RoyPurdue U (USA)
    Hans-Joachim WunderlichStuttgart U (DE)