About
I am passionate about information processing and computation in complex natural and artificial distributed systems. This includes, for instance, cortical areas, behaviour in swarms, locally distributed computing, or deep neural networks, to name just a few. My interests revolve particularly around plasticity, emergence, self-organization, self-construction, and collective behaviour, and how such dynamics can lead to stable computations within non-linear dynamical systems. My goal is to identify the algorithms & data structures that not only explain these properties, but also follow from first-order principles with as few parameters as possible. In addition, I am enthusiastic about parallel scientific computing and computer graphics.
Currently, I focus on emergence of hierarchical and modular computational structures in distributed message-passed systems. Moreover, I conduct research on optimality, self-organization, and recursive self-construction of (scale-free) cortical data structures, in particular those involved in spatial navigation.
Bio
After working on a PhD in Computer Science and as a postdoc in the Neuroscientific System Theory (NST) group at TU Munich, Germany, I spent almost three years as research scientist at the Bosch Center for Artificial Intelligence. Since March 2021 I'm a postdoc/researcher in Yasser Roudi's SPINOr Group at the Kavli Institute for Systems Neuroscience (KISN) in beautiful Trondheim, Norway.
Throughout my career, I have always loved working on the intersection of computational neuroscience, computer science, and neuromorphic computing. In this context, I aimed my research primarily at those neural computations that are involved in spatial navigation and perception. Moreover, I worked on technical robotic solutions that operate in the real-world, and led a team researching and developing a novel perception framework for manipulation robotics. Given my strong interest in teaching and education, I also conceived and implemented a mentoring program that includes reverse mentoring for higher management.
Please contact me if you are interested in a detailed bio.
Contact
External links
List of Publications
2021- A. Kupcsik, M. Spies, A. Klein, M. Todescato, N. Waniek, P. Schillinger, M. Buerger. Supervised Training of Dense Object Nets Using Optimal Descriptors for Industrial Robotic Applications. AAAI 2021.
- L. Rozo , M. Guo , A. G. Kupcsik , M. Todescato, P. Schillinger, M. Giftthaler, M. Ochs, M. Spies, N. Waniek, P. Kesper, M. Burger. Learning and Sequencing of Object-Centric Manipulation Skills for Industrial Tasks. IROS'2020.
- N. Waniek. Transition scale-spaces: A computational theory for the discretized entorhinal cortex. Neural Computation 32(2):330-394, Feb 2020,. MIT Press. DOI: 10.1162/neco_a_01255.
- Leonel Rozo, Andras Kupcsik, Meng Guo, Marco Todescato, Philipp Schillinger, Nicolai Waniek, Markus Giftthaler, Mathias Bürger. Fast Learning and Sequencing of Object-centric Manipulation Skills. R:SS 2019, June 22 2019, Freiburg.
- M. Spies, M. Todescato, H. Becker, P. Kesper, N. Waniek, M. Guo. Bounded Suboptimal Search with Learned Heuristics for Multi-Agent Systems. Proceedings of the AAAI Conference on Artificial Intelligence, Volume 33, pages 2387-2394, 2019. DOI: 10.1609/aaai.v33i01.33012387
- N. Waniek. Locally distributed spatial navigation in a scale-space model for grid cells. PhD thesis, 2018.
- N. Waniek. Hexagonal Grid Fields Optimally Encode Transitions in Spatiotemporal Sequences. Neural Compututation 30(10):2691-2725, Oct 2018. MIT Press. DOI: 10.1162/neco_a_01122.
- Z. Tayeb, N. Waniek, J. Fedjaev, L. Rychly, N. Ghaboosi, C. Widderich, C. Richter, J. Braun, M. Saveriano, G. Cheng, and J. Conradt. gumpy: A Python Toolbox Suitable for Hybrid Brain-Computer Interfaces. Journal of Neural Engineering, 15(6), pp-065003, 2018. DOI: 0.1088/1741-2552/aae186.
- N. Waniek, and J. Conradt. Grid Cells as Transition Encoders. Poster presented at Bernstein Conference on Computational Neuroscience, Göttingen, 2017.
- Preprint: N. Waniek. Multi-Transition Systems: A theory for neural spatial navigation. 2017.
- Preprint: N. Waniek, J. von Stetten, and J. Conradt. Graph cuts for asynchronous event-based vision sensors. 2017.
- M. Mulas, N. Waniek, and J. Conradt. Hebbian plasticity realigns grid cell activity with external sensory cues in continuous attractor models. Front Comput Neurosci, 10:13, Feb 2016.
- Preprint: N. Waniek, E. Berzs, and J. Conradt. Data Structures for Locally Distributed Routing. 2016.
- N. Waniek, J. von Stetten, and J. Conradt. Event-based graph cuts, 2016. Poster presented at Neurocomputing Systems Workshop, Frauenwörth, 2016.
- N. Waniek, J. Biedermann, and J. Conradt. Cooperative SLAM on small mobile robots. In 2015 IEEE International Conference on Robotics and Biomimetics (ROBIO), pages 1810–1815, Dec 2015.
- M. Mulas, N. Waniek, and J. Conradt. Exploiting grid cell properties for robotic spatial navigation. Poster presented at BCCN Retreat, Tutzing, 2015.
- N. Waniek, M. Mulas, and J. Conradt. Self-organization of grid cell networks. Poster presented at Bernstein Conference on Computational Neuroscience, Heidelberg, 2015.
- N. Waniek, S. Bremer, and J. Conradt. Real-time anomaly detection with a growing neural gas. In Artificial Neural Networks and Machine Learning – ICANN 2014, volume 8681 of Lecture Notes in Computer Science, pages 97–104. Springer International Publishing, 2014.
- R. Araújo, N. Waniek, and J. Conradt. Development of a dynamically extendable spinnaker chip computing module. In Artificial Neural Networks and Machine Learning – ICANN 2014, volume 8681 of Lecture Notes in Computer Science, pages 821–828. Springer International Publishing, 2014.
- M. Mulas, N. Waniek, and J. Conradt. Neuromorphic architecture for robotic spatial navigation. Poster presented at Bernstein Conference on Computational Neuroscience, Göttingen, 2014.
- N. Waniek, M. Mulas, and J. Conradt. Grid cell realignment based on idiothetic head direction cues. Poster presented at Bernstein Conference on Computational Neuroscience, Göttingen, 2014.
- N. Waniek, C. Denk, and J. Conradt. GRIDMAP – from brains to technical implementations. Poster presented at Bernstein Conference on Computational Neuroscience, Tübingen, 2013.
- N. Waniek, C. Denk, and J. Conradt. GRIDMAP – from brains to technical implementations. Poster presented at Bernstein Conference on Computational Neuroscience, Tübingen, 2013.
- N. Waniek and J. Conradt. From brains to technical implementations. 2013. Poster presented at BCCN Sparks Workshop, Tutzing, 2013.
- N. Waniek. Biologically Inspired Model for Visually Driven Navigation. Diploma Thesis, 2012.
