Curriculum Vitae

Education

• 2022 - present: PhD Student at Friedrich-Alexander-Universität Erlangen-Nürnberg, Department Hardware-Software-Co-Design
• 2019 - 2021: M.S. in Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg
  • Thesis: Energy Efficient Deployment of Deep Learning Applications on Restricted Hardware using Deep Compression (in cooperation with Fraunhofer IIS Nuremberg)
  • Relevant Coursework: Artificial Intelligence, Deep Learning, Operating Systems Design, Distributed Systems, Virtual Machines
• 2015 - 2019: B.S. in Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg
  • Thesis: Android Inter Process Communication Fuzzing
  • Relevant Coursework: Computer Graphics, Embedded Systems, Hyperperformance Computing

Work Experience

• 2022 - present: Researcher at Friedrich-Alexander-Universität Erlangen-Nürnberg, Department Hardware-Software-Co-Design and Fraunhofer IIS, Group Efficient AI
  • Automated design, compression, and deployment of deep neural networks (DNNs) on Cortex-M microcontrollers for image and time series classification and regression tasks
  • On-device training of fully quantized DNNs on Cortex-M microcontrollers
  • Design, implementation, and maintenance of three demonstrators for live audio classification, object detection, and on-device training of DNNs on Cortex-M microcontrollers
• 2020 - 2022: Research Assistant at Fraunhofer IIS, Group “Self-Learning Systems” and “Machine Learning & Validation”, Nuremberg
  • Compression of deep neural networks for deployment on restricted hardware
  • Duties includes: Updates and improvements to template
  • Evaluation of different parameter reduction, quantization, and graph optimization techniques for deep neural networks
• 2015 - 2020: Working Student at Siemens AG, Erlangen
  • Design, development and integration of a webservice into the open, cloud based IoT system Mindsphere
  • Continuous delivery and maintenance of the webservice as a tool used by Mindsphere´s system test and quality assurance departments
  • Implementation of an Android application and a web UI for the webservice
  • Implementation of automated UI tests based on the Selenium framework

Service

• I have served as a reviewer for NeurIPS’24 ‘25 and ‘26, ICLR’25, TMLR, ACM Transactions on Embedded Computing Systems, IEEE Internet of Things Journal, and Springer Journal of Real-Time Image Processing.

Publications

EmbeddedAD: Efficient Anomaly Detection on SoCs via Vector-Quantized Reverse Distillation

Deutel, M., Marchl, A., Plinge, A., Hannig, F., Teich, J., (2026). EmbeddedAD: Efficient Anomaly Detection on SoCs via Vector-Quantized Reverse Distillation. In: IEEE 39th International System-on-Chip Conference (SOCC).

Early-Exit Forecasting of Deep Neural Networks on Energy-Harvesting Edge Devices

Sixdenier, P., Deutel, M., Wildermann, S., Teich, J., (2026). Early-Exit Forecasting of Deep Neural Networks on Energy-Harvesting Edge Devices. In: ECML PKDD Conference, at the 7th Workshop on IoT, Edge, and Mobile for Embedded Machine Learning.

Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollers

Schubert, J., Kasper, M., Linke, M., Herzog, B., Deutel, M., Plinge, A., Seuss, D., & Mutschler, M. (2026). Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollers. arXiv: 2608.03589.

PrototypeNAS: Rapid Design of Deep Neural Networks for Microcontroller Units

Deutel, M., Geis, S., Plinge, A. (2026). PrototypeNAS: Rapid Design of Deep Neural Networks for Microcontroller Units. Machine Learning and Knowledge Discovery in Databases. Research Track. ECML PKDD 2026. Lecture Notes in Computer Science, vol 16941. Springer, Cham.

Efficient Network Inference via Hardware-Aware Architecture Search, Model Pruning & Quantization

Heublein, L., Deutel, M., Plinge, A., & Ott, F. (2026). Efficient Network Inference via Hardware-Aware Architecture Search, Model Pruning & Quantization. arXiv:2606.23210.

Recent Trends in Edge AI: Efficient Design, Training and Deployment of Machine Learning Models

Deutel, M., Mallah M., Wissing J., Scheele, S. (2026). Recent Trends in Edge AI: Efficient Design, Training and Deployment of Machine Learning Models. Charting the Intelligence Frontiers – Edge AI Systems Nexus.

Unsupervised Learning of Variational Autoencoders on Cortex-M Microcontrollers

Deutel, M., Plinge, A., Seuß, D., Mutschler, C., Hannig, F., & Teich, J. (2025). Unsupervised Learning of Variational Autoencoders on Cortex-M Microcontrollers. IEEE 18th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC).

Early-Exit Neural Architecture Search for Energy-Harvesting Edge Computing

Sixdenier, P., Deutel, M., & Teich, J. (2025). Early-Exit Neural Architecture Search for Energy-Harvesting Edge Computing. IEEE 18th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC).

Multi-Objective Bayesian Optimization with Reinforcement Learning for Edge Deployment of DNNs on Microcontrollers

Deutel, M., Kontes, G., Mutschler, C., & Teich, J. (2025). Multi-Objective Bayesian Optimization with Reinforcement Learning for Edge Deployment of DNNs on Microcontrollers. Proceedings of the Genetic and Evolutionary Computation Conference Companion (GECCO).

Combining Multi-Objective Bayesian Optimization with Reinforcement Learning for TinyML

Deutel, M., Kontes, G., Mutschler, C., & Teich, J. (2025). Combining Multi-Objective Bayesian Optimization with Reinforcement Learning for TinyML. ACM Transactions on Evolutionary Learning and Optimization.

Fused-Layer CNNs for Memory-Efficient Inference on Microcontrollers

Deutel, M., Hannig F., Mutschler, C., & Teich, J. (2024). Fused-Layer CNNs for Memory-Efficient Inference on Microcontrollers. In Workshop on Machine Learning and Compression @ NeurIPS.

Energy-Efficient AI on the Edge

Witt, N., Deutel, M., Schubert, J., Sobel, C., Woller, P. (2024). Energy-Efficient AI on the Edge. In: Mutschler, C., Münzenmayer, C., Uhlmann, N., Martin, A. (eds) Unlocking Artificial Intelligence. Springer, Cham.

On-Device Training of Fully Quantized Deep Neural Networks on Cortex-M Microcontrollers

Deutel, M., Hannig, F., Mutschler, C., & Teich, J. (2024). On-Device Training of Fully Quantized Deep Neural Networks on Cortex-M Microcontrollers, in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

microYOLO: Towards Single-Shot Object Detection on Microcontrollers

Deutel, M., Mutschler, C., & Teich, J. (2023). microYOLO: Towards Single-Shot Object Detection on Microcontrollers. In ECML PKDD Conference 2023, at the 4th Workshop on IoT, Edge, and Mobile for Embedded Machine Learning.

Energy-efficient Deployment of Deep Learning Applications on Cortex-M based Microcontrollers using Deep Compression

Deutel, M., Woller, P., Mutschler, C., & Teich, J. (2023, March). Energy-efficient Deployment of Deep Learning Applications on Cortex-M based Microcontrollers using Deep Compression. In MBMV 2023; 26th Workshop (pp. 1-12). VDE.