Connectionists: 2 Postdoc positions in modeling dendritic integration and synaptic plasticity

[Walter Senn]

Dear Colleagues, dear all,

We are inviting applications for 2 postdoc positions (for at least 2 
years, earliest start April 1, 2016) in computational neuroscience at 
the University of Bern, Switzerland.

Our research is devoted to models of synaptic plasticity and dendritic 
integration that serve as a computational paradigm for learning in 
neuronal networks. We have recently suggested to consider “Learning by 
the dendritic prediction of somatic spiking” (Neuron 2014) that 
assigns an intrinsic computational task to neurons and dendrites. This 
concept will be extended by including dendritic nonlinearities and 
considering networks of such neuronal prediction elements. The 2 
positions are:

1) Postdoc for modeling dendritic integration. We have shown how the 
error-backpropagation algorithm translates to a dendritic tree that 
displays NMDA-spikes in individual dendritic branches (PLoS Comp Biol 
2016). Along these ideas a biophysical model of dendritic integration 
will be worked out that makes also contact to Bayesian integration.

2) Postdoc for modeling synaptic plasticity and learning in networks. 
The neurons as intrinsic prediction elements will be connected to form 
a network that mimics various learning tasks and that is particularly 
suited for imitation learning.

You will profit from generous travel funding and from collaborations 
in- and outside the Human Brain Project, and you will also enjoy the 
beautiful city and surrounding of Switzerland’s capital. Ideal 
candidates have a strong background in computational neuroscience and 
machine learning.

Please send CV, letter of motivation, and publication list by March 11 
to Walter Senn (senn at pyl.unibe.ch) and Sabine Herzog 
(herzog at pyl.unibe.ch).

With best wishes
Walter Senn

PS: For recent work of our group see 
http://www.physio.unibe.ch/~senn/neuroscience.aspx