Implementation of Backtracking Line Search in Deep Neural Networks - Theory and Practice

Explore the theory and practical implementation of backtracking line search in deep neural networks in this 1 hour 18 minute conference talk by Dr. Tuyen Truong from VinAI. Gain insights from Dr. Truong's diverse mathematical background spanning pure and applied mathematics, with a focus on deep learning applications. Learn how his experience as an amateur Go player influenced his interest in AI. Discover the potential of backtracking line search to optimize neural network training and performance. Delve into both theoretical foundations and hands-on implementation strategies for incorporating this technique into deep learning workflows.