Sunny talks to Dr. Dharmendra Modha, who has pioneered neuromorphic computing at IBM for nearly three decades. He brought the concept of neuromorphic engineering to the wider technical community through his TrueNorth chips and has continued to push field forward. His latest advance, the North Pole inference system, demonstrates remarkable energy efficiency: 75 times more efficient than traditional GPUs. In this episode, he discusses both the architecture and axioms behind this breakthrough, discusses the 3 and 8 billion parameter models his team has demonstrated, and shares insights on the future of brain-inspired computing. After the interview, discussion follows with Giulia and Ralph.
Articles and Papers
- A million spiking-neuron integrated circuit with a scalable communication network and interface
- Neural inference at the frontier of energy, space, and time
- Breakthrough low-latency, high-energy-efficiency LLM inference performance using NorthPole
From the discussion
- Networks of spiking neurons: The third generation of neural network models
- The role of single neurons in information processing
- Neuromorphic implementation of orientation hypercolumns
- Expandable networks for neuromorphic chips
- Neurogrid: A mixed-analog-digital multichip system for large-scale neural simulations
- Probabilistic synaptic weighting in a reconfigurable network of VLSI integrate-and-fire neurons
- A multichip neuromorphic system for spike-based visual information processing
- Integrating with neurons
- Saccades and the quick phase of nystagmus
- VIDEO: The future of Artificial Intelligence: 3D Silicon Brain
- Dendrocentric learning for synthetic intelligence
- A comprehensive data-driven model of cat primary visual cortex
- Light speed machine learning inference on the edge
Episode Credits
Producer/Writer: Sunny Bains
Co-hosts: Sunny Bains and Giulia D’Angelo
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.
