Scientists Use Noise to Make Brain-Like Computers Smarter
New research shows random noise can make AI chips more efficient and powerful.
A paper on neuromodulation in neuromorphic control design investigates how spiking excitability can be modulated by intrinsic noise. Using a simplified memristive neuron model, it shows that intrinsic noise plays distinct functional roles depending on feedback pathways: noise in fast excitatory feedback shapes the nature of excitability and modulates sensitivity to external/internal stimuli, while noise in slow inhibitory feedback provides internal stimuli that can trigger stochastic spiking. The authors expect these findings to guide the design of neuromorphic systems that harness intrinsic noise to exploit beneficial stochastic phenomena such as stochastic resonance.
- Noise in brain-like computer chips can be beneficial, not just a problem.
- Different types of noise in a neuron's feedback loops have different effects on how it fires.
- This research could lead to more efficient AI hardware that uses noise to improve performance.
Why It Matters
This could lead to AI chips that are faster, more energy-efficient, and better at handling real-world unpredictability.