This modern and self-contained book offers a clear and accessible introduction to the important topic of machine learning with neural networks. In addition to describing the mathematical principles of the topic, and its historical evolution, strong connections are drawn with underlying methods from statistical physics and current applications within science and engineering. Closely based around a well-established undergraduate course, this pedagogical text provides a solid understanding of the key aspects of modern machine learning with artificial neural networks, for students in physics, mathematics, and engineering. Numerous exercises expand and reinforce key concepts within the book and allow students to hone their programming skills. Frequent references to current research develop a detailed perspective on the state-of-the-art in machine learning research.
- Accessible and self-contained approach to the theoretical foundations of machine learning, and its modern applications in science and engineering
- Suitable for teaching advanced undergraduate or graduate courses on this topic for science, mathematics and engineering students
- Numerous end of chapter exercises expand and reinforce key concepts
Conditions of Use
This book is licensed under a Creative Commons License (CC BY-NC-SA). You can download the ebook Machine Learning with Neural Networks for free.
- Title
- Machine Learning with Neural Networks
- Subtitle
- An Introduction for Scientists and Engineers
- Publisher
- Cambridge University Press
- Author(s)
- Bernhard Mehlig
- Published
- 2021-10-28
- Edition
- 1
- Format
- eBook (pdf, epub, mobi)
- Pages
- 241
- Language
- English
- ISBN-10
- 1108494935
- ISBN-13
- 9781108494939
- License
- CC BY-NC-SA
- Book Homepage
- Free eBook, Errata, Code, Solutions, etc.
Acknowledgements Contents 1 Introduction 1.1 Neural networks 1.2 McCulloch-Pitts neurons 1.3 Activation functions 1.4 Asynchronous updates 1.5 Summary 1.6 Further reading I Hopfield networks 2 Deterministic Hopfield networks 2.1 Pattern recognition 2.2 Hopfield networks and Hebb's rule 2.3 The cross-talk term 2.4 One-step error probability 2.5 Energy function 2.6 Summary 3 Stochastic Hopfield networks 3.1 Stochastic dynamics 3.2 Order parameters 3.3 Mean-field theory 3.4 Critical storage capacity 3.5 Beyond mean-field theory 3.6 Correlated and non-random patterns 3.7 Summary 3.8 Further reading 4 The Boltzmann distribution 4.1 Convergence of the stochastic dynamics 4.2 Monte-Carlo simulation 4.3 Simulated annealing 4.4 Boltzmann machines 4.5 Restricted Boltzmann machines 4.6 Summary 4.7 Further reading II Supervised learning 5 Perceptrons 5.1 A classification problem 5.2 Iterative learning algorithm 5.3 Gradient descent for linear units 5.4 Classification capacity 5.5 Multi-layer perceptrons 5.6 Summary 5.7 Further reading 6 Stochastic gradient descent 6.1 Chain rule and error backpropagation 6.2 Stochastic gradient-descent algorithm 6.3 Preprocessing the input data 6.4 Overfitting and cross validation 6.5 Adaptation of the learning rate 6.6 Summary 6.7 Further reading 7 Deep learning 7.1 How many hidden layers? 7.2 Vanishing and exploding gradients 7.3 Rectified linear units 7.4 Residual networks 7.5 Outputs and energy functions 7.6 Regularisation 7.7 Summary 7.8 Further reading 8 Convolutional networks 8.1 Convolution layers 8.2 Pooling layers 8.3 Learning to read handwritten digits 8.4 Coping with deformations of the input distribution 8.5 Deep learning for object recognition 8.6 Summary 8.7 Further reading 9 Supervised recurrent networks 9.1 Recurrent backpropagation 9.2 Backpropagation through time 9.3 Vanishing gradients 9.4 Recurrent networks for machine translation 9.5 Reservoir computing 9.6 Summary 9.7 Further reading III Learning without labels 10 Unsupervised learning 10.1 Oja's rule 10.2 Competitive learning 10.3 Self-organising maps 10.4 K-means clustering 10.5 Radial basis functions 10.6 Autoencoders 10.7 Summary 10.8 Further reading 11 Reinforcement learning 11.1 Associative reward-penalty algorithm 11.2 Temporal difference learning 11.3 Q-learning 11.4 Summary 11.5 Further reading
