My first paper! Will be presenting this work as a poster at the HiLD workshop at ICML and CCN - please reach out if you want to chat :)
Reposted from
Ezekiel Williams
🧵 1/5: Psyched to share a new preprint! In it, we show neural networks can approximate wide classes of functions/dynamical systems when synaptic weights are random and only unit biases are learned! This should be of interest to fellow neuroscientists, and to the machine learning crowd.
Comments