Freiburg/ Berlin, September 14, 2018
For decades, neuroscientists have used artificial neural networks simulated by computers to better understand the functioning of the human brain.
The award winner and his research
Wulfram Gerstner studies spiking neural networks (SNN), in which neurons communicate with each other by means of short pulses (action potentials or ’spikes‘), similar to real biological neural networks.
Gerstner is a pioneer. With his theoretical modelling, he proposed hypotheses early on, which could only later be experimentally confirmed. His research on synaptic plasticity („spike-timing dependent plasticity“ (STDP)), the adaptability of the contact points between nerve cells, is outstanding. This property is at the basis of neuronal learning. In one of his first studies, Gerstner contributed decisively to the understanding of the activity-dependent development of a biological nervous system using the auditory system of the barn owl as an example.
Recent work of his research group investigates the interface between learning processes in artificial neural networks and learning in biological neural networks, using an artificial network that learns to compose music as an example.
Wulfram Gerstner was one of the first scientists to introduce benchmarks for neuron models in computational neuroscience by comparing their response to a specific input stimulus to a well-defined set of experimental data. As a result, it was possible to simplify models of neural networks considerably. „Despite their reduced complexity, these neuron models are surprisingly more accurate than some of the more sophisticated biophysical models – important food for thought for future scientific approaches in modelling,“ says Ad Aertsen, chairman of the jury of the Valentino Braitenberg Award for Computational Neuroscience.
Wulfram Gerstner is the author of the book „Spiking Neuron Models. Single Neurons, Populations, Plasticity” (2002), which has become an indispensable textbook in computational neuroscience.
The award will be conferred at the Bernstein Conference on September 26, 2018 at TU Berlin. Following the award ceremony, Gerstner will give the Valentino Braitenberg Lecture, in which he will present highlights of his research.
Exclusive to Journalists
Media representatives are cordially invited to attend the award ceremony and the Valentino Braitenberg Lecture. Please register at the reception of the Bernstein Conference in the main building of the TU Berlin. If you wish, you can also interview the award winner. Please contact one of the two press representatives.
Valentino Braitenberg Award for Computational Neuroscience
The Valentino Braitenberg Award for Computational Neuroscience was created to commemorate the renowned Tübingen neuroscientist Valentino Braitenberg (1926-2011). Every two years, the Bernstein Network is presenting the award to an outstanding scientist whose research has contributed significantly to our understanding of brain functions.
In the spirit of Valentino Braitenberg’s research, a particular focus is placed on theoretical studies. Valentino Braitenberg is considered a pioneer in the modern field of computational neuroscience and has made a significant contribution to the development of the field of biological cybernetics. His research has also inspired robotics and artificial intelligence.
The award is donated by the Autonomous Province of Bolzano-Alto Adige.
Bernstein Network Computational Neuroscience
The Bernstein Network is a research network in the field of computational neuroscience; this field brings together experimental approaches in neurobiology with theoretical models and computer simulations. The network started in 2004 with a funding initiative of the Federal Ministry for Education and Research (BMBF) to develop and interconnect regional research structures in computational neuroscience throughout Germany and to promote the transfer of theoretical insight into clinical and technical applications. In this context, computational neuroscience joins experimental approaches in neurobiology with theoretical models and computer simulations.
The network is named after the German physiologist and biophysicist Julius Bernstein (1839-1917). After more than 10 years of funding by the Federal Ministry, the Bernstein Network consists of more than 200 research groups. Since 2016, the continuity of the network is sustained by the non-profit association Bernstein Network Computational Neuroscience.