Gated Recurrent Neural Networks with Weighted Time-Delay Feedback

Explore gated recurrent neural networks with weighted time-delay feedback in this 28-minute conference talk delivered by Soon Hoe Lim from Nordita, KTH Royal Institute of Technology and Stockholm University. Presented at the Fourth Symposium on Machine Learning and Dynamical Systems, hosted by the Fields Institute on July 11, 2024, this lecture delves into advanced concepts in neural network architecture. Gain insights into the integration of time-delay feedback mechanisms within gated recurrent structures, and understand their implications for enhancing the performance and capabilities of machine learning models in dynamic system applications.