Deep Learning and Inference Using Models with Low Precision Synapses and Binary Unit Activations

Organization
Office of the Director of National Intelligence (ODNI)
Reference Code
ICPD-2021-14
How to Apply

Create and release your Profile on Zintellect – Postdoctoral applicants must create an account and complete a profile in the on-line application system.  Please note: your resume/CV may not exceed 2 pages.

Complete your application – Enter the rest of the information required for the IC Postdoc Program Research Opportunity. The application itself contains detailed instructions for each one of these components: availability, citizenship, transcripts, dissertation abstract, publication and presentation plan, and information about your Research Advisor co-applicant.

Additional information about the IC Postdoctoral Research Fellowship Program is available on the program website located at: https://orise.orau.gov/icpostdoc/index.html.

If you have questions, send an email to ICPostdoc@orau.org.  Please include the reference code for this opportunity in your email. 

Application Deadline
2/26/2021 6:00:00 PM Eastern Time Zone
Description

Research Topic Description, including Problem Statement:

Deep learning has revolutionized the field of machine learning. However, current approaches to deep neural network training require backpropagation of errors with high precision. This poses a challenge for training deep neural networks on future generation, low-power edge computing platforms under the constraint of low-precision (possibly binary) weights and binary unit activations. Recent research on binary neural networks (BNNs) and spiking neural networks (SNNs) offers hope that a solution to this problem can be found. However, this remains an open problem. Approaches may involve simulators, field-programmable gate arrays (FPGAs), and/or neuromorphic hardware.

Example Approaches:

  • “EventProp: Backpropagation for Exact Gradients in Spiking Neural Networks” arXiv:2009.08378
  • “Training Binary Neural Networks with Real-to-Binary Convolutions” arXiv:2003.11535

Relevance to the Intelligence Community:

The Intelligence Community (IC) will increasingly rely on edge computing platforms to detect patterns of interest in sensor data. Developing algorithms for training and inference on future generation, low-power edge computing platforms will ensure that the IC is able to take full advantage of these platforms.

Key Words: Deep Learning, Machine Learning, Artificial Intelligence, Neuromorphic, Low Precision, Event Based Computing, Neural Networks, SNN, BNN

Qualifications

Postdoc Eligibility

  • U.S. citizens only
  • Ph.D. in a relevant field must be completed before beginning the appointment and within five years of the application deadline
  • Proposal must be associated with an accredited U.S. university, college, or U.S. government laboratory
  • Eligible candidates may only receive one award from the IC Postdoctoral Research Fellowship Program

Research Advisor Eligibility

  • Must be an employee of an accredited U.S. university, college or U.S. government laboratory
  • Are not required to be U.S. citizens
Eligibility Requirements
  • Citizenship: U.S. Citizen Only
  • Degree: Doctoral Degree.
  • Discipline(s):
    • Communications and Graphics Design (2 )
    • Computer, Information, and Data Sciences (17 )
    • Earth and Geosciences (20 )
    • Engineering (27 )
    • Environmental and Marine Sciences (15 )
    • Life Health and Medical Sciences (46 )
    • Mathematics and Statistics (11 )
    • Nanotechnology (1 )
    • Other Non-S&E (2 )
    • Other Physical Sciences (12 )
    • Physics (16 )
    • Social and Behavioral Sciences (27 )