- ICNS Home
- Projects
- Open PhD Projects
- A Neuromorphic Framework for Event-Based DNNs using Minifloats
A Neuromorphic Framework for Event-Based DNNs using Minifloats
Supervisors:
Description:
Researching a novel neuromorphic framework that performs proven deep learning algorithms (to solve real world problems) using event-based signal processing. Biological cortex, which intrinsically computes with discrete-time events by using spiking neurons, exhibit complex yet stereotypical network architectures that support rich dynamics. Developing the computing models for our neuromorphic framework using low precision formats. Recent 8-bit foating-point (minifoat) representations used by DNNs have achieved marginal equivalent accuracy to FP32 foating point precision over dikerent tasks and datasets while providing orders of magnitude reduction in silicon area and power consumption. Minifoats are ideal candidates for neuromorphic computing. This framework will have a set of minifoat neuromorphic computing models that include a versatile three compartment spiking neuron model. This neuromorphic framework will use event-based DNNs that only compute changes, which is significantly more effcient than updating all the layers each time. By combining the event-based processing and the minifoat together, this framework will be capable of achieving marginal differences in accuracy compared to the corresponding frame based DNNs with full precision foating point number over different tasks and datasets while with orders of magnitudes reduction in computation resouces in terms of memory footprint and foating point operations per second (FLOPS).
About ICNS
Partnership
Research
Projects
-
- Task-Driven Model Evaluation in Large-Scale Spiking Neural Networks
- A Neuromorphic Ferroelectric field-effect Ultra-Scaled Chip for Spiking Neural Networks
- Event Based Wavefront Sensing Modalities
- Physics-Based Encoding for Spiking Neural Networks
- Neuromorphic Computational Imaging
- Defining Performance Metrics for Closed Loop Event Based Imaging Systems
- A Neuromorphic Framework for Event-Based DNNs using Minifloats
- A RISC-V instruction set architecture (ISA) extensions for neuromorphic computing using minifloats
- Astrometry with Event-based Vision Sensors
- Automatic Evaluation of Bushfire Risk via Acoustic Scene Analysis
- Bio-inspired Sensors for Space Situational Awareness
- Building a Neuromorphic Auditory Pathway for Sensing the Surrounding Environment
- Cold Start Astrometry for High-Precision Airspace and Space Objects Tracking with Neuromorphic Cameras
- Control Systems Inspired by Insect Central Pattern Generators that can Adapt to Dynamic Environments.
- Design of Neuromorphic Spiking Neural Networks for Real-Time Processing
- Enhanced Maritime Situational Awareness with Neuromorphic Cameras
- Environmental Situational Awareness using Neuromorphic Vision Sensors and IMU-based SLAM
- Fault Tolerant Distributed Swarm Intelligence using Neuromorphic Computing and Local Learning Principles
- Honey Bee Waggle Dance Detection via Neuromorphic Engineering
- Integrated Circuit Design for Event-based Vision Sensors
- Low-Power Acoustic Ecological Monitoring in Remote Areas using Machine Learning and Neuromorphic Engineering
- Neuromorphic Computing in Extreme Environments
- Neuromorphic Cyber Security at the Edge
- Neuromorphic Engineering for Acoustic Aerial Drone Detection in Visually Obscured Environments
- Machine Learning-Based Tool for Therapists to Monitor Speech Progress in Late Talkers
- Machine Learning for Automated Child Reading Assessment and Intervention
- Underwater Acoustic Drone Detection via Neuromorphic Models of Marine Mammal Audition
Resources
Master's Program
Opportunities
Events
News
Contact