In this episode Sunny talks to Prof Emre Neftci, Director of the Neuromorphic Software Ecosystems group at the Peter Grünberg Institute (PGI). He and his colleagues, part of the Jülich Research Centre in Germany, think about how neurons can be trained and organized to learn in an efficient and brain-inspired way. You’ll hear about his work in making backpropagation compatible with spiking neural networks, dealing with device variability, and one- and few-shot learning. Discussion follows with Giulia and Ralph.
Articles and Papers
- Meta-learning spiking neural networks with surrogate gradient descent
- HyperSpike: Hyperdimensional computing for more efficient and robust spiking neural networks
- Neuro-inspired computing for next-gen AI: Computing model, architectures and learning algorithms
From the Discussion
- Representation ensembling for synergistic lifelong learning with quasilinear complexity
- HOTS: A hierarchy of event-based time-surfaces for pattern recognition
- Experts weigh impact of Prophesee-Qualcomm deal
- Targeted muscle reinnervation for real-time myoelectric control of multifunction artificial arms
- Online few-shot gesture learning on a neuromorphic processor
Episode Credits
Producer/Writer: Sunny Bains
Co-hosts: Sunny Bains and Giulia D’Angelo
Commentator: Ralph Etienne-Cummings
Editor: Rose Gotto
Audio Production: Taylor Marvin, Coupe Studios Music and Sound Design
Music: 3 of Diamonds by Matt Harris
For EETimes.com
This episode is produced in conjunction with EETimes. Podcast reproduced with permission.
