Recap and Closing Remarks: Neuropunk revolution and energy consumption
summary
Argues that spiking neuromorphic hardware can cut AI energy use by orders of magnitude, demonstrates a hardware model of a rat spinal cord using memristive synapses and two plastic neurons that self-organize an output pattern matching rehabilitative gait after one minute versus 40 weeks biologically, and claims an implantable, microwatt device and robotic controllers are feasible. Timeline for practical systems is 5–7 years based on current prototypes and minimal neuron counts.
AI-written summary · about
chapters (9)
- 0:00Neuropunk closing intro
- 1:51Energy in digital neural nets
- 7:13Brain rules for efficiency
- 10:52Neuroscience inspiration
- 14:48Hardware AI without software
- 16:55Spinal cord tech future
- 18:00Spinal tech timeline Q&A
- 19:25Spiking neural network revolution
- 20:08Closing remarks
related talks
01h 57mIntroduction to Neuromorphic Computing
41mBrain-inspired computational primitives for reliable neuromorphic intelligence
01h 16mRethinking Computing with Neuro-Inspired Learning: Algorithms to Hardware
42mOn the Link Between Energy & Information for the Design of Neuromorphic Systems
43mKeeping Moore's law alive: Neuromorphic computing
29mDeep Learning, Neuroscience and the future of AI
18mNews from the Neuro Geeks
21mOrganoid Computing – Can Nature Save AI?