Towards adaptive spiking label propagation

Conference · Sun Jul 01 04:00:00 EDT 2018

DOI:https://doi.org/10.1145/3229884.3229897· OSTI ID:1468094

Hamilton, Kathleen ^[1]; Schuman, Catherine ^[1]

---

1. ORNL

Graph algorithms are a new class of applications for neuromorphic hardware. Rather than adapting deep learning and standard neural network approaches to a low-precision spiking environment, we use spiking neurons to analyze undirected graphs (e.g., the underlying modular structure). While fully connected spin glass implementations of spiking label propagation have shown promising results on graphs with dense communities, identifying sparse communities remains difficult. This work focuses on steps towards an adaptive spike-based implementations of label propagation, utilizing sparse embeddings and synaptic plasticity. Sparser embeddings reduce the number of inhibitory connections, and synaptic plasticity is used to simultaneously amplify spike responses between neurons in the same community, while impeding spike responses across different communities. We present results on identifying communities in sparse graphs with very small communities.

• Research Organization: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: USDOE
• OSTI ID: 1468094
• Resource Type: Conference paper/presentation
• Conference Information: International Conference on Neuromorphic Systems - Knoxville, Tennessee, United States of America - 7/23/2018-7/26/2018
• Country of Publication: United States
• Language: English