BSC researcher Jordi Torres publishes the first book in Spanish on deep reinforcement learning

21 May 2021

The Emerging Technologies for artificial intelligence group leader at the Barcelona Supercomputing Center (BSC), Jordi Torres, has published the book "Introduction to deep reinforcement learning. Theory and practice in Python". The manual is intended to be an initiation guide for those who want to know the theoretical foundations of the area. At the same time, from the Python implementation and detailed explanation of all the algorithms presented in the book, working knowledge of the fundamentals is provided.

It is the first book in Spanish on this area. "The complexity of this field of AI and the lack of books in Spanish made it difficult for researchers to enter this area," Torres said.

The Emerging Technologies for artificial intelligence group leader has divided the book into four parts. The first deals with the key elements in reinforcement learning. The second presents the fundamental algorithms, which are the basis of current algorithms. In the third part, the BSC researcher reviews the basic concepts of deep learning and introduces a practical version through the PyTorch library. In the fourth part, it is presented how deep learning has revived interest in reinforcement learning and how these algorithms can be implemented with new libraries that are changing the scene. The book concludes with a closing chapter that appeals to the responsibility of the engineers who use this type of technology.

"This book is an excellent introduction to the theory and applications of reinforcement learning and should serve as a solid foundation for your advancement on this topic. I hope this book not only serves as an introduction to the community and established issues, but that is also useful to go further and deepen in this area, since there are many opportunities in research and applications in industry ", Torres stressed.

Although reinforcement learning is often described as a special part of machine learning, the reality is that it is driving some of the most important improvements for society. Different recent research articles suggest the application of reinforcement learning in healthcare, education systems or smart cities, among many other areas.

Jordi Torres, who is also a professor in the Department of Computer Architecture at the Universitat Politècnica de Catalunya, has published several books, including some on big data, cloud computing and artificial intelligence.