Tag
The author describes building a spiking neural network engine that initially failed the NARMA-10 benchmark, but by applying heterogeneous wire delays from neuroscience, it achieved usable memory depth and a 15x computational efficiency advantage over continuous nets on a recognition task.
A developer shares Helix-AGI, a continuously-running cognitive agent using a physics-based memory retrieval system that integrates recency, structural importance, and semantic proximity via an entropic gravity equation and Euler-Lagrange dynamics, without tuning separate weights.