Lectures related to Automatic differentiation

Covers tensors, loss functions, autograd, and convolutional layers in deep learning.

Covers the analysis of scalar fields, including divergence, gradient, and Laplacian.

Discusses recurrent neural networks, their training challenges, and solutions like LSTMs and GRUs.

Covers the evaluation methods for natural language generation systems, including metrics and human assessments of generated text quality.

Covers state space models and their expressivity compared to transformers, focusing on attention mechanisms and computational efficiency.