Includes bibliographical references and index.

Deep learning basics. Scope of this textbook -- Models for data analysis -- Building blocks of neural networks -- Optimization of network parameters -- Mastering model building -- Standard architectures of deep networks. Revisiting the terminology -- Fully-connected networks: improving the classic all-rounder -- Convolutional neural networks and analysis of image-like data -- Recurrent neural networks: time series and variable input -- Graph networks and convolutions beyond Euclidean domains -- Multi-task learning, hybrid architectures, and operational reality -- Introspection, uncertainties, objectives. Interpretability -- Uncertainties and robustness -- Revisiting objective functions -- Deep learning advanced concepts. Beyond supervised learning -- Weakly-supervised classification -- Autoencoders: finding and compressing structures in data -- Generative models: data from noise -- Domain adaptation, refinement, unfolding -- Model independent detection of outliers and anomalies -- Beyond the scope of this textbook.

"A core principle of physics is knowledge gained from data. Thus, deep learning has instantly entered physics and may become a new paradigm in basic and applied research. This textbook addresses physics students and physicists who want to understand what deep learning actually means, and what is the potential for their own scientific projects. Being familiar with linear algebra and parameter optimization is sufficient to jump-start deep learning. Adopting a pragmatic approach, basic and advanced applications in physics research are described. Also offered are simple hands-on exercises for implementing deep networks for which python code and training data can be downloaded."-- Provided by publisher.

Mode of access: World Wide Web.

System requirements: Adobe Acrobat reader.