steve abreu

prof_pic.jpg

Bernoulliborg 407

9747 AG Groningen

The Netherlands

s.abreu@rug.nl

I am interested in interdisciplinary research in and around computer science and artificial intelligence: nature-inspired computing, understanding human cognition, and advancing artificial intelligence.

My PhD is on non-digital computing theory in the MINDS research group and the CogniGron research center, funded by the European Post-Digital research network. I am developing concepts and methods for programming and interfacing with unconventional computers. I am working with analog neuromorphic hardware (spikes!), photonic computing systems, and other physical systems that can compute and/or learn.

In my free time I am a hobby photographer, like to read, travel, and go to music and art events.

meeting requests

To schedule a meeting, please use this page to book a time in my calendar.

supervision (bachelor/master)

I am available to supervise bachelor and master projects at RUG in fall 2022. If you are interested in working on a project related to machine learning or neural networks in novel neuromorphic hardware systems, please send me an email.

Topic ideas that I would find particularly interesting:

  • learning in neural networks using local learning rules for bio-plausibility and efficient hardware implementation: equilibrium propagation and learning algorithms using dendritic processing are particularly promising.
  • evolving modular components in recurrent neural network architectures.
  • exploring different local learning rules in recurrent neural networks.
  • autonomous multi-task learning for recurrent neural networks (e.g. using conceptors).
  • neurosymbolic programming for neuromorphic hardware: merging machine learning with “classical” programming.

Take a look at our group’s website for more information on the kind of research we do, and this page for more information on doing your thesis project in our group.

news

Jun 2022 Back in Groningen after three months at the Institute of Neuroinformatics (ETH/UZH in Zurich).
May 2022 Talk on physical reservoir computing to the CogniGron research center (slides).
Mar 2022 Internal Post-Digital presentation of progress on programming spiking neuromorphic hardware.
Feb 2022 Presentation on my PhD topic to the CogniGron center of cognitive materials (slides).

selected publications

  1. Hands-on reservoir computing: a tutorial for practical implementation
    Cucchi, Matteo,  Abreu, Steven, Ciccone, Giuseppe, Brunner, Daniel, and Kleemann, Hans
    Neuromorphic Computing and Engineering 2022
  2. Automated architecture design for deep neural networks
    Abreu, Steven
    ArXiv Aug 2019