Thursday, March 14, 2013

[Comp-neuro] Post-doc position at NTU, Singapore on Role of Active Dendrites in Boosting Information Processing Capacity of Spiking Neural Networks

*Job Title:  Research Fellow (Post Doc)*

 

*Area of expertise: Computational Neuroscience, Machine Learning*

 

*Project Description*

The selected person will be working on algorithms for a project titled “Neuro-inspired Reconfigurable Processor: Circuits with Emergent Structure”. This project aims to develop strategies for efficient recognition and classification systems by taking inspiration from neuroscience. In particular, the role of nonlinear dendrites in neurons to boost the computational power of a network will be studied using single and multi-compartment dendritic models. New learning rules that involve changing morphologies of neurons (as opposed to weight plasticity) will be developed which is inspired by structural plasticity observed in human brains. Both supervised and unsupervised learning rules need to be developed and the effect of these rules in eliciting emergent properties in recurrent networks need to be analyzed. The possible relationship between weight plasticity and structural plasticity will also be explored. Application of these novel dendritic cells and morphological learning to liquid state machines will be studied. The work is to be done in collaboration with faculty from Georgia Tech, USA and Institute of Neuroinformatics, Zurich.

 

*Requirements from Candidate*

This position requires a candidate with a Ph.D. degree who is proficient in Machine Learning and Computational Neuroscience. In particular he/she must have the following skills:

•             Experienced in computational neuroscience and more specifically reduced spiking neuron models (e.g. leaky integrate and fire, Izhikevich etc), synaptic models and spike based learning models (e.g. Spike Timing Dependent Plasticity). Knowledge of single and multi-compartment modelling of dendrites will be an added advantage.

•             Experienced in Machine Learning using Neural Networks. Familiar with concepts of supervised and unsupervised learning, batch training, generalization theory etc.

•             Proficient in MATLAB coding.

•             Familiarity with liquid state machine computation and statistical learning theory (e.g. independent/ principal component analysis) is an added bonus.

The person should be able to lead research activity and guide graduate students in a team. He/She should be able to communicate clearly and be experienced in writing technical articles (journal/conference) papers.

 

*Application Procedure*

Send your CV with two of your most relevant publications to arindam.basu@ntu.edu.sg.

 

*Enquiries*

·         Asst. Prof. Arindam Basu

School of Electrical & Electronic Engineering
Nanyang Technological University
S2-B2C-84, 50 Nanyang Avenue
Singapore 639798

 

arindam.basu@ntu.edu.sg

 

*Start Date*

Earliest start date is April 1, 2013.

 

 

 

Regards,

 

Arindam

 

http://www3.ntu.edu.sg/home/arindam.basu/

 



CONFIDENTIALITY:This email is intended solely for the person(s) named and may be confidential and/or privileged.If you are not the intended recipient,please delete it,notify us and do not copy,use,or disclose its content.

Towards A Sustainable Earth:Print Only When Necessary.Thank you.

No comments:

Post a Comment