Friedemann Zenke
| Friedemann Zenke | |
|---|---|
| File:Friedemannn Zenke Portrait FMI 2020.jpgFriedemannn Zenke Portrait FMI 2020.jpg Friedemann Zenke at the FMI in 2020 | |
| Born | |
| 🏳️ Citizenship | German |
| 💼 Occupation | |
| 🌐 Website | https://www.zenkelab.org |
Friedemann Zenke is a junior group leader at the Friedrich Miescher Institute for Biomedical Research[1] and holds the position of Assistant Professor at the University of Basel, Switzerland.[2] He is best known for his theoretical work on spiking neural networks, synaptic plasticity, and learning.
Early career and education
Zenke started his academic career studying Physics at the University of Bonn. His interest in computational neuroscience led him to the École polytechnique fédérale de Lausanne (EPFL), Switzerland. Working in the laboratory of Prof. Wulfram Gerstner, he completed his Ph.D. in 2014, focusing on "Memory formation and recall in recurrent spiking neural networks."[3]
Zenke conducted post-doctoral research at Stanford University, California. From 2015 to 2017, he worked with Prof. Surya Ganguli, publishing research on continual learning.[4] In 2017, he moved to the University of Oxford as a Sir Henry Wellcome Postdoctoral Fellow.[5] Until 2019, he worked in the lab of Prof. Tim P. Vogels at the Department of Physiology, Anatomy and Genetics, University of Oxford, and the Centre for Neural Circuits and Behaviour.[6]
Research and contributions
Zenke's research lies at the intersection of neuroscience and machine learning. His work has been featured by Quanta Magazine, specifically in the area of neuromorphic algorithms.[7] This article was also translated into German by Spektrum.[8] His contributions to the field include the following topics:
- Neural Network Learning and Plasticity
This involves the development and exploration of models that can learn from and adapt to their input. Notably he worked on continual learning and co-authored the article "Continual learning through synaptic intelligence."[9] His work further comprises memory formation in spiking neural networks including publications such as "Diverse synaptic plasticity mechanisms orchestrated to form and retrieve memories in spiking neural networks,"[10] and "The temporal paradox of Hebbian learning and homeostatic plasticity."[11]
- Spiking Neural Networks and Neuromorphic Computing
Zenke has done considerable research in the area of spiking neural networks, a type of artificial neural network that attempts to mimic the neurophysiology of biological neurons more closely. His work in this area includes the development of learning algorithms for these networks and their application to neuromorphic computing, a field that aims to implement neurobiological architectures into computing. Relevant publications of his include "Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to Spiking Neural Networks",[12] "Superspike: Supervised learning in multilayer spiking neural networks",[13] and "Brain-inspired learning on neuromorphic substrates".[14]
- Applications of Deep Learning in Neuroscience
Zenke has contributed to the application of deep learning techniques in the field of neuroscience, with the goal of providing better frameworks for understanding neural mechanisms and processing. His publications such as "A deep learning framework for neuroscience"[15] and "Visualizing a joint future of neuroscience and neuromorphic engineering"[16] reflect this aspect of his work.
- Development of Auryn - A Neural Network Simulator
Zenke developed Auryn, a simulator for spiking neural networks with plastic synapses. Unlike other simulators, Auryn emphasizes simulation speed for efficient modelling of small to medium-sized networks over extended periods. It leverages optimized designs for shared memory systems, using Message Passing Interface (MPI) for parallel execution and Single Instruction, Multiple Data (SIMD) operations for faster computations[17]. Zenke's Auryn prioritizes modularity and extensibility, facilitating easy integration of new neuron or synapse types. Besides, he maintains a comprehensive wiki to support community learning and contributions. Zenke offers Auryn as a free open-source software under the GNU General Public License showcasing his commitment to accessible, collaborative development in computational neuroscience[18]
Selected publications
- Zenke, F., Poole, B., & Ganguli, S. (2017). Continual learning through synaptic intelligence. In International conference on machine learning (pp. 3987-3995). PMLR.
- Neftci, E. O., Mostafa, H., & Zenke, F. (2019). Surrogate gradient learning in spiking neural networks: Bringing the Power of Gradient-based optimization to Spiking Neural Networks. IEEE Signal Processing Magazine, 36(6), 51-63.
- Vogels, T. P., Sprekeler, H., Zenke, F., Clopath, C., & Gerstner, W. (2011). Inhibitory plasticity balances excitation and inhibition in sensory pathways and memory networks. Science, 334(6062), 1569-1573.
- Cramer, B., Billaudelle, S., Kanya, S., Leibfried, A., Grübl, A., Karasenko, V., ... & Zenke, F. (2022). Surrogate gradients for analog neuromorphic computing. Proceedings of the National Academy of Sciences, 119(4), e2109194119.
- Halvagal, M. S., and Zenke, F. (2023). The Combination of Hebbian and Predictive Plasticity Learns Invariant Object Representations in Deep Sensory Networks. Nature Neuroscience 26, 1906–1915.
References
- ↑ "FMI Group Leader Friedemann Zenke". Retrieved 2023-09-11.
- ↑ url=https://philnat.unibas.ch/de/personen/friedemann-zenke/
- ↑ "PhD Thesis: Memory formation and recall in recurrent spiking neural networks" (PDF).
- ↑ "Neural Dynamics and Computation Lab". ganguli-gang.stanford.edu. Retrieved 2023-05-30.
- ↑ "Sir Henry Wellcome Postdoctoral Fellowship: Role of complex synaptic dynamics for learning and memory". Retrieved 2023-09-11.
- ↑ "Director Candidates 2020". www.cnsorg.org. Retrieved 2023-05-30.
- ↑ Whitten, Allison (17 February 2022). "AI Overcomes Stumbling Block on Brain-Inspired Hardware". Quanta Magazine.
- ↑ "Neuromorphe Computer: Vorbild Gehirn". Spektrum (in Deutsch). 2022.
- ↑ Zenke, Friedemann; Poole, Ben; Ganguli, Surya (2017). "Continual Learning Through Synaptic Intelligence". arXiv:1703.04200 [cs.LG].
- ↑ Zenke, Friedemann; Agnes, Everton J.; Gerstner, Wulfram (2015-04-21). "Diverse synaptic plasticity mechanisms orchestrated to form and retrieve memories in spiking neural networks". Nature Communications. 6 (1): 6922. Bibcode:2015NatCo...6.6922Z. doi:10.1038/ncomms7922. ISSN 2041-1723. PMC 4411307. PMID 25897632.
- ↑ Zenke, Friedemann; Gerstner, Wulfram; Ganguli, Surya (2017-04-01). "The temporal paradox of Hebbian learning and homeostatic plasticity". Current Opinion in Neurobiology. Neurobiology of Learning and Plasticity. 43: 166–176. doi:10.1016/j.conb.2017.03.015. ISSN 0959-4388. PMID 28431369.
- ↑ Neftci, Emre O.; Mostafa, Hesham; Zenke, Friedemann (November 2019). "Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-Based Optimization to Spiking Neural Networks". IEEE Signal Processing Magazine. 36 (6): 51–63. Bibcode:2019ISPM...36f..51N. doi:10.1109/MSP.2019.2931595. ISSN 1053-5888. Unknown parameter
|s2cid=ignored (help) - ↑ Zenke, Friedemann; Ganguli, Surya (2018). "Super Spike: Supervised Learning in Multilayer Spiking Neural Networks". Neural Computation. 30 (6): 1514–1541. doi:10.1162/neco_a_01086. PMC 6118408. PMID 29652587. Retrieved 2023-05-30.
- ↑ Zenke, Friedemann; Neftci, Emre O. (May 2021). "Brain-Inspired Learning on Neuromorphic Substrates". Proceedings of the IEEE. 109 (5): 935–950. doi:10.1109/JPROC.2020.3045625. ISSN 0018-9219. Unknown parameter
|s2cid=ignored (help) - ↑ Richards, Blake A.; Lillicrap, Timothy P.; Beaudoin, Philippe; Bengio, Yoshua; Bogacz, Rafal; Christensen, Amelia; Clopath, Claudia; Costa, Rui Ponte; de Berker, Archy; Ganguli, Surya; Gillon, Colleen J.; Hafner, Danijar; Kepecs, Adam; Kriegeskorte, Nikolaus; Latham, Peter (November 2019). "A deep learning framework for neuroscience". Nature Neuroscience. 22 (11): 1761–1770. doi:10.1038/s41593-019-0520-2. ISSN 1546-1726. PMC 7115933 Check
|pmc=value (help). PMID 31659335. - ↑ Zenke, Friedemann; Bohté, Sander M.; Clopath, Claudia; Comşa, Iulia M.; Göltz, Julian; Maass, Wolfgang; Masquelier, Timothée; Naud, Richard; Neftci, Emre O.; Petrovici, Mihai A.; Scherr, Franz; Goodman, Dan F.M. (February 2021). "Visualizing a joint future of neuroscience and neuromorphic engineering". Neuron. 109 (4): 571–575. doi:10.1016/j.neuron.2021.01.009. hdl:10044/1/88104. ISSN 0896-6273. PMID 33600754 Check
|pmid=value (help). Unknown parameter|s2cid=ignored (help) - ↑ Zenke, Friedemann; Gerstner, Wulfram (2014-09-11). "Limits to high-speed simulations of spiking neural networks using general-purpose computers". Frontiers in Neuroinformatics. 8: 76. doi:10.3389/fninf.2014.00076. ISSN 1662-5196. PMC 4160969. PMID 25309418.
- ↑ "start [Auryn simulator]". fzenke.net. Retrieved 2023-05-30.
External links
This article "Friedemann Zenke" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Friedemann Zenke. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
