AutoML-Zero

This video explores AutoML-Zero, an evolutionary search for machine learning programs. These programs are initially empty with Setup, Predict, and Learn functions that can access scalar, vector, and matrix memory addresses. Through fitness evaluation and mutation, these programs evolve to use gradient descent, dropout-like operations, and ReLU activation functions! Thanks for watching, Please Subscribe! Paper Links: AutoML-Zero: https://ift.tt/2Q4IQCl Github Repo AutoML-Zero: https://ift.tt/2Iyj0Cq The Evolved Transformer: https://ift.tt/2IAdYFw Hierarchical Representations for Efficient Architecture Search: https://ift.tt/2AA2JIh Exploring Randomly Wired Neural Networks for Image Recognition: https://ift.tt/2v9Xi4y