Circuits and Systems for Artificial Intelligence

Artificial Intelligence (AI) applications require the use more and more advanced algorithms for extracting meaningful information from increasingly (and sometimes incredibly) large set of data. While the algorithmic part has been recently seen significant advances (as for instance through the adoption of Deep or Convolutional Neural Networks), it sometimes comes at the cost of high computational complexity which hinders their straightforward implementability. This activity aims at advancing in this direction by:

  1. proposing architectures to reduce the computational cost of AI algorithms either by solely hardware design changes or by joint hardware-algorithm co-design. Examples are the use of weights with largely reduced precision in DNNs or minimization of data transfer bringing the computation at the edge of the cloud;
  2. Changing the algorithms for processing big data to largely reduce memory requirements and hardware complexity. An example in this direction is the use of suitably modified streaming principal component analysis algorithms.

ERC Sector:

  • PE7_4 (Micro and nano) systems engineering
  • PE7_7 Signal processing
  • PE7_11 Components and systems for applications


  • Neural network hardware
  • Artificial intelligence

Research groups