Mastering Deep Learning with Java

More Information
Learn
  • Set up your Java environment for deep learning using deeplearning4j
  • Understand the deep learning architectures for classification and prediction
  • Implement state-of-art recent attention-based models for natural language processing
  • Interpret and debug a deep learning model by visualizing its hidden states and activations
  • Learn about building deep reinforcement learning agents
  • Understand the limitations of current deep learning models and approaches to overcome them
About

Neural-net based models with multiple layers have demonstrated to be efficient compact function approximators that generalize beyond the training data. Research in deep learning is largely done in Python, causing enterprises where Java dominates to lag behind in leveraging state-of-art deep learning models in applications. This book attempts to fill that void.

Mastering Deep Learning with Java brings the utilization of the next-generation capability of deep neural nets and finding patterns in your datasets using powerful deep learning techniques. You will develop the capabilities to develop complex neural network models using Deeplearning4j, one of Java's popular and preferred libraries for deep learning. The book would teach you how you can design optimized neural networks in the areas of computer vision, natural language processing, and computer games.

This book covers the workhorse models of deep learning starting with simple feedforward neural nets, convolution neural nets, sequence models, to the more recently popular attention based models (GPT-2, BERT etc.)

By the end of this book, you will be well equipped to design, develop and deploy efficient, enterprise-grade deep learning models from scratch using Java.

Features
  • Leverage Java’s deep learning capabilities to train enterprise-grade neural network models
  • Use deeplearning4j to train CNNs, RNNs, attention-based models and more
  • Fine-tune the performance of your deep learning models and deploy them to production
Page Count 590
Course Length tbc
ISBN 9781789132960
Date Of Publication 23 Jan 2020