Connectionists: Postdoc position in large-scale neural modeling -- Eliasmith lab

[Peter Blouw]

In brief, the Eliasmith lab at the Centre for Theoretical Neuroscience has
a position available for a 2-year postdoc.  The focus is on large-scale
spiking neural models that exhibit interesting behaviors and run on
supercomputing infrastructure.  "Interesting behaviour" is broadly defined,
including vision, motor control, decision making, any variety of learning,
audition, language processing, and so on.

Full details below:



*SOSCIP/IBM Canada Post-Doctoral Fellow (PDF) Research Scientist*



Launched in 2012, SOSCIP was founded by seven Ontario universities and IBM
Canada Limited as a collaborative research consortium, with a mandate to
bring together academic researchers and small- and medium-sized companies
to drive innovation using state-of-the-art advanced computing and big data
analytics technologies, commercial outcomes for social and economic
development in Ontario.  New investments from IBM Canada and other
stakeholders have enabled SOSCIP to expand its computing platforms and
resources to increase capacity and add new members.  With the addition of
our three new academic partners, membership in the consortium has more than
doubled over the last two years to 16 organizations.



The IBM Canada Research & Development Centre is expanding, and we are
seeking a post-doctoral fellow who has a PhD in Theoretical or
Computational Neuroscience or a related field, with strong programming
skills and experience in high-performance computing environments.



This role will support a research project led by Dr. Chris Eliasmith in the
Centre for Theoretical Neuroscience at the University of Waterloo. Dr.
Eliasmith's lab constructs state-of-the-art, large-scale, neuron-level
models of a wide variety of behaviours, recently publishing what remains
the world's largest functional brain model, Spaun, in Science.  These
models are simulated in the software Nengo, which is able to run models on
a wide variety of hardware platforms, or 'backends', including neuromorphic
hardware.  Recently, a backend for BlueGene/Q has been developed, allowing
Nengo models, including Spaun, to run on the supercomputers housed at
SOSCIP.  The successful candidate will focus on scaling up models,
including Spaun, to take full advantage of these hardware resources.
Scaling will include developing theory and implementation of new
behaviours, as well as increasing the biological fidelity of Nengo models.



A strong publication record, with demonstrated applications of neural-level
models to behaviour (e.g. vision, motor control, memory, decision making,
etc.), neuron-level compartmental modeling, various applications of
learning (deep learning, reinforcement learning, STDP, etc.), and
large-scale modeling are greatly desired.  Familiarity with Python, MPI,
and Javascript would be beneficial.





Education and Experience

-       PhD in Theoretical or Computational Neuroscience, or a related field

-      strong programming skills and experience in high-performance
computing environments

-      experience working as part of a research team





Work Location:   Waterloo, Ontario



Term of Contract:   2-years maximum



IBM is committed to creating a diverse environment and is proud to be an
equal opportunity employer. All qualified applicants will receive
consideration for employment without regard to race, color, religion,
gender, gender identity or expression, sexual orientation, national origin,
genetics, disability, age, or veteran status.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20160131/b3235bf6/attachment.html>