Written by an expert, Krishnendu Chaudhury, This book, “Math and Architectures of Deep Learning,” is perfectly balanced as it perfectly combines mathematical theory and implementation in Python and PyTorch. That makes it a unique resource that fills the gap between research papers and practical applications, as the authors provide the readers with an idea of how to easily adjust, solve issues with, and fine-tune deep learning architectures.
Every aspect of neural networks, from linear algebra to vector calculus and multivariate statistics, is explained so the readers have a good grounding in the mathematical paradigms supporting neural networks. The book “Math and Architectures of Deep Learning” brings readers up to speed with the fundamental concepts and designs of deep learning models so they can apply the findings from the latest studies. By having side-by-side context explanations of the theory and the actual code offered in the downloadable Jupyter notebooks, the readers will be well-equipped to tackle deep learning.
Math and Architectures of Deep Learning assist data scientists with the necessary knowledge to address the process of model selection, training, and assessment. Thus, by opening the ‘black box,’ readers will not only know how their code works but will also gain enough knowledge to turn poorly performing models into well-optimized ones and apply state-of-the-art structures.
Math and Architectures of Deep Learning Table of Contents:
- An overview of machine learning and deep learning
- Vectors, matrices, and tensors in machine learning
- Classifiers and vector calculus
- Linear algebraic tools in machine learning
- Probability distributions in machine learning
- Bayesian tools for machine learning
- Function approximation: How neural networks model the world
- Training neural networks: Forward propagation and backpropagation
- Loss, optimization, and regularization
- Convolutions in neural networks
- Neural networks for image classification and object detection
- Manifolds, homeomorphism, and neural networks
- Fully Bayes model parameter estimation
- Latent space and generative modelling, autoencoders, and variational autoencoders
Who is this course for?
- Persons with a background in data science, machine learning engineers, and researchers intending to use TensorFlow who have some Python computer programming language skills and a basic understanding of algebraic and calculus facts.
