Franz Scherr

PhD Candidate at Institute of Theoretical Computer Science

About

I am a PhD Candidate at the Institute of Theoretical Computer Science, University of Technology Graz. Under supervision of Professor Wolfgang Maass, I design and investigate algorithms to advance learning mechanism in brain-inspired artificial neural networks. In particular, my interests include topics of Computational Neuroscience, Machine- and Deep Learning.

In addition, I am advising Trever in questions concerning Machine Learning.

[CV]

You can reach out to me at: [firstname] [dot] [lastname] [at] [tugraz] [dot] [at]

Publications

  • A solution to the learning dilemma for recurrent networks of spiking neurons
    Bellec G*, Scherr F*, Subramoney A, Hajek E, Salaj D, Legenstein R, Maass W
    Nature Communications 2020

  • One-shot learning with spiking neural networks
    Scherr F, Stöckl C, Maass W
    bioRxiv 2020 | under review

  • Eligibility Traces provide a data-inspired alternative to backpropagation through time
    Bellec G*, Scherr F*, Hajek E, Salaj D, Subramoney A, Legenstein R, Maass W
    NeurIPS 2019 | Real neurons and hidden units workshop

  • Slow processes of neurons enable a biologically plausible approximation to policy gradient
    Subramoney A*, Bellec G*, Scherr F*, Hajek E, Salaj D, Legenstein R, Maass W
    NeurIPS 2019 | Biological and artificial RL workshop

  • Reservoirs learn to learn
    Subramoney A, Scherr F, Maass W
    arXiv 2019

  • Neuromorphic Hardware learns to learn
    Bohnstingl T*, Scherr F*, Pehle C, Meier K, Maass W
    Frontiers in Neuroscience | Neuromorphic Engineering 2019

  • Biologically inspired alternatives to backpropagation through time for learning in recurrent neural nets
    Bellec G*, Scherr F*, Hajek E, Salaj D, Legenstein R, Maass W
    arXiv 2019

    • * equal contribution and team work

  • Spike-based agents for Multi-armed bandits (Master's thesis 2018, Information and Computer Engineering)
    Advisor: Prof. Wolfgang Maass
    [PDF]

  • Gradient-based optimization of AMEA parameters (Bachelor's thesis 2018, Physics)
    Advisor: Prof. Wolfgang von der Linden
    [PDF]

  • Automated Security Proofs for Symmetric Ciphers (Bachelor's thesis 2016, Telematik)
    Advisor: Maria Eichlseder
    [PDF]