Etienne Mueller

Welcome to my personal website!

I am a postdoctoral researcher at the University of Melbourne, Australia. My work spans computational neuroscience, AI, and neuromorphic computing, where I focus on bridging biological principles with advanced machine learning models.

This site serves as a hub for my professional journey, current interests, and ongoing explorations in technology and research. In the future, it will evolve into a playground for experimental projects, where I’ll share tools, ideas, and updates related to my work.

For now, feel free to browse my CV, explore my Google Scholar profile for publications, or check out my open-source contributions on GitHub. Thank you for visiting, and stay tuned for more!

 

Resume

Professional Experience

May 2023 - present
University of Melbourne

Postdoctoral Researcher Artificial Intelligence & Computational Neuroscience

  • Developing a neural network growth algorithm to create biologically-inspired memory cells for more efficient recurrent neural networks, using ML for simulations of brain imaging data across different developmental stages using JAX
  • Running deep convolutional neural networks on a Slurm-based HPCwith up to 4xH100 GPUs per node for automated segmentation of synchrotron brain imaging data, reducing the need for manual annotation by a factor of 10

Oct. 2022 - Mar. 2023
Flowers-Software GmbH

AI Engineer

  • Established the AI research department at a seed-financed startup, deploying deep learning infrastructure from scratch
  • Developed a TensorFlow-based information extraction workflow on AWS to identify recurring positions on invoices

Oct. 2021 - Sep. 2022
Technical University of Munich (TUM), Germany

Technical Advisor

  • Led the technical coordination of a pilot case for an EU-funded (Horizon 2020) project (SHOP4CF)
  • Creation of modular tools for autonomous factories for Industry 4.0

Oct. 2018 - Sep. 2022
Infineon Technologies AG

AI Researcher

  • Research in neuromorphic computing and spiking neural networks, leading to 11 first- and second-author publications
  • Developed a TensorFlow-based toolbox for converting conventional to spiking neural networks, which was subsequently used in a research project to reduce the simulation time of hardware components in neuromorphic systems by half
  • Research cooperation with the Technical University of Munich

Mar. 2018 - Sep. 2018
BMW AG

Component Manager

  • Technical supervision of cooling water pumps for electric and combustion vehicles for BMW, Mini and Rolls Royce
  • Requirement engineering and long-term testing of different models in cooperation with Bosch and Continental

Aug. 2012 - Aug. 2017
e-gnition Hamburg e.V. @ TUHH

Formula Student Team

Developer Driverless Actuator Technology (2016 - 2017)
  • 1st Place Formula Student Driverless: Autonomous Design
  • 3rd Place Formula Student Driverless: Overall
Division Manager Business Plan (2014 - 2016)
  • Special Award for educational video "How to Business Plan" at Formula Student Hungary
President & Team Captain (2013 - 2014)
  • Special Award for Ecological Design by Magna Steyr
Division Manager Aerodynamics (2012 - 2013)

Jan. 2015 - Sep. 2016
Slive

Co-founder & CEO

  • Developed smart wearable devices and location-based algorithms for hands-free data use in industrial environment
  • Secured the Nissen Foundation Start-Up Grant 3,000€ to support early-stage product development and business growth

Oct. 2013 - Sep. 2014
Institute of Aircraft Production, TUHH

Research Assistant

  • Manufacturing of components via a six-axis industrial robot and CAM software

Sep. 2010 - Sep 2012
Lischke Consulting

Assistant Consultant

  • Supported various corporate restructuring projects

Education

Jun. 2023 - Jun. 2023
University of Waterloo, Canada

Nengo Summer School

  • Nengo summer school on large-scale brain modelling and neuromorphic computing

Oct. 2018 - Sep. 2022
Technical University of Munich (TUM), Germany

Ph.D. in Computer Science

  • Thesis on the conversion of conventional to spiking neural networks for energy-efficient neuromorphic computation
  • Research of novel biologically-inspired approaches for convolutional, recurrent and transformer architectures for natural language processing and pattern recognition with varying large datasets

Apr. 2015 - Aug. 2017
Technical University of Hamburg (TUHH), Germany

M.Sc. in Product Development, Materials and Production

  • Thesis on the development of autonomous racing vehicles to sense and act under time constraints
  • Awarded Best Autonomous Design at Formula Student Germany
  • Awarded the Incentive Prize of the Technical University of Hamburg endowed with 1,500€

Sep. 2015 – Feb. 2016
Institut Catholique d’Arts et Métiers Nantes (ICAM), France

Semester Abroad

  • Thesis on designing tools for workflow optimization of a waste utilization plant using CAD

Oct. 2009 – Dec. 2014
Technical University of Hamburg (TUHH), Germany

B.Sc. in Mechanical Engineering

  • Thesis on measuring and increasing the accuracy and repeatability of industrial robots using MATLAB

Open Source Projects

High-Performance Zebrafish (HPZ)

– A Python and bash toolkit designed to automate recurring brain imaging data tasks on a Slurm-based HPC setup
– End-to-end pipeline that consolidates multiple manual steps for loading, preprocessing, and detecting neurons and spikes in microscopy data into a single automated process, reducing manual intervention and error by a factor of five
– Automated setup for new users to easily work with zebrafish brain imaging data, improving onboarding efficiency

Convert2SNN

– A TensorFlow-based library that converts conventionally trained neural networks with continuous activation functions to spiking neural networks, with minimal to no performance loss, to estimate energy consumption in neuromorphic systems
– Supports key spike encoding techniques, including rate, population, and temporal coding, with the ability to estimate spike counts for efficiency evaluation, reducing the need for extensive hardware simulations during development

For the full list see: Github

Teaching Experience

Mar. 2019 - Sep 2021

Cognitive Systems

  • Covering topics such as cognition, senses and perceptions, biological inspired computational models and neurorobotics
  • Creating and grading of exams for over 400 students

Oct. 2019 - Sep. 2021

Thesis Supervision in Deep Learning and Spiking Neural Networks

  • Master Thesis (2021): Performance of Time to First Spike Encoded Spiking Neural Networks
  • Research Internship (2021): Conversion of Analog to Spiking Transformer Networks
  • Master Thesis (2021): Conversion of Analog LSTM-based Recurrent Neural Networks
  • Master Thesis (2021): Conversion of Analog GRU-based Recurrent Neural Networks
  • Research Internship (2020): Carla as Open Source Platform for Analyzing and Evaluating Autonomous Driving
  • Master Thesis (2020): Converting Analog to Spiking Convolutional Neural Networks for Object Detection
  • Master Thesis (2019): Semantic Segmentation of Integrated Circuit Layout Images

Publications

Reverse Engineering Neural Connectivity: Mapping Neural Activity Data to Artificial Neural Networks for Synaptic Strength Analysis

E. Mueller, W. Qin, in 8th International Conference on Information Technology (InCIT), Chonburi, Thailand and Kanazawa, Japan, 2024, (accepted).

Neural Oscillations for Energy-Efficient Hardware Implementation of Sparsely Activated Deep Spiking Neural Networks,

E. Mueller, S. Klimaschka, D. Auge, A. Knoll, in Association for the Advancement of Artificial Intelligence (AAAI) Practical DL, Online (Vancouver, Canada), 2022, pp. 1-7.

Exploiting Inhomogeneities of Subthreshold Transistors as Populations of Spiking Neurons

E. Mueller, D. Auge, A. Knoll, in International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), Online (Fuzhou, China), 2022, pp. 1-8.

Spiking Transformer Networks: A Rate Coded Approach for Processing Sequential Data

E. Mueller, V. Studenyak, D. Auge, A. Knoll, in 7th Int. Conference on Systems and Informatics (ICSAI), Online (Jiaxing, China), 2021, pp. 1-5.

Minimizing Inference Time: Optimization Methods for Converted Deep Spiking Neural Networks

E. Mueller, J. Hansjakob, D. Auge, A. Knoll, in International Joint Conference on Neural Networks (IJCNN), Online (Shenzen, China), 2021, pp. 1-8.

A Survey of Encoding Techniques for Signal Processing in Spiking Neural Networks

D. Auge, J. Hille, E. Mueller, A. Knoll, Neural Processing Letters, vol. 53, issue 6, pp. 4693-4710, Dec 2021.

Normalization Hyperparameter Search for Converted Spiking Neural Networks

E. Mueller, D. Auge, A. Knoll, in Bernstein Computational Neuroscience Conference, Online (Berlin, Germany), 2021, P 8.

Hand Gesture Recognition in Range-Doppler Images Using Binary Activated Spiking Neural Networks

D. Auge, J. Hille, E. Mueller, A. Knoll, in IEEE International Conference on Automatic Face and Gesture Recognition, Online (Jodhpur, India), 2021, pp. 1-7.

End-to-end Spiking Neural Network for Speech Recognition Using Resonating Input Neurons

D. Auge, J. Hille, F. Kreutz, E. Mueller, A. Knoll, in 30th International Conference on Artificial Neural Networks (ICANN), Online (Bratislave, Slovakia), 2021, pp. 245-256.

Faster Conversion of Analog to Spiking Neural Networks by Error Centering

E. Mueller, J. Hansjakob, D. Auge, in Bernstein Computational Neuroscience Conference, Online (Berlin, Germany), 2020, P 146.

Hand Gesture Recognition using Hierarchical Temporal Memory on Radar Sequence Data

D. Auge, P. Wenner, E. Mueller, in Bernstein Computational Neuroscience Conference, Online (Berlin, Germany), 2020, P 3.

Resonate-and-Fire Neurons as Frequency Selective Input Encoders for Spiking Neural Networks

Daniel Auge, E. Mueller, Chair of Informatics, TUM, Munich, Technical Report TUM-I2083. 2020

For the full list see: Google Scholar

Programming

Python

TensorFlow

bash

Git

Swift & Xcode

MATLAB

Java

C++

Languages

German

French

English

Spanish

Chinese

Contact

How Can I Help You?