In this project, we leverage neuromorphic computing algorithms and hardware to address four important needs in machine learning (ML) at edge devices:

  1. Energy-Efficient Computing: One of the major bottlenecks of ML in edge devices is their severe energy constraints.

  2. Incremental Learning: In most of the edge devices data arrives in form of a stream rather than batches, therefore an ML model is required to learn continuously and incrementally upon the arrival of each sample data.

  3. Capture Temporal Information: data streams produced by edge devices often exhibit temporal patterns and dependencies, which are important to be captured to distinguish the relationship between input features over time.

  4. Drift Tolerance: edge devices normally operate in non-stationary environments, and are prone to concept drifts that are induced by appearance and/or disappearance of features and/or classes in the incoming data stream.

Here, we use the on-chip neuromorphic-enhanced edge computing (Neuro-Edge) approach as an alternative for power-hungry off-chip deep learning methods for ML at edge devices. We will leverage Intel's neuromorphic chip, Loihi, to prototype the Neuro-Edge system.

Project Repository:


Student Contributors:

  • Brenden Chavis

  • Praful Chunchu