A memory-efficient, gradient-free zeroth-order (derivative-free) optimizer designed to solve the "Curse of Dimensionality" in Black-Box optimization and memory-constrained Machine Learning. It provides an O(log D) gradient estimation approach that can successfully train Neural Networks without ever calculating analytical derivatives or Backprop
python machine-learning deep-learning black-box-optimization derivative-free-optimization gradient-free-optimization zeroth-order-optimization neural-network-training high-dimensional-optimization backpropagation-alternative denoised-gradient-estimation
-
Updated
Apr 26, 2026 - Python