Something to read

We continue studying different reinforcement learning methods.

more...

---------------------

1. Neural networks made easy
2. Neural networks made easy (Part 2): Network training and testing
3. Neural networks made easy (Part 3): Convolutional networks
4. Neural networks made easy (Part 4): Recurrent networks
5. Neural networks made easy (Part 5): Multithreaded calculations in OpenCL
6. Neural networks made easy (Part 6): Experimenting with the neural network learning rate
7. Neural networks made easy (Part 7): Adaptive optimization methods
8. Neural networks made easy (Part 8): Attention mechanisms
9. Neural networks made easy (Part 9): Documenting the work
10. Neural networks made easy (Part 10): Multi-Head Attention
11. Neural networks made easy (Part 11): A take on GPT
12. Neural networks made easy (Part 12): Dropout
13. Neural networks made easy (Part 13): Batch Normalization
14. Neural networks made easy (Part 14): Data clustering
15. Neural networks made easy (Part 15): Data clustering using MQL5
16. Neural networks made easy (Part 16): Practical use of clustering
17. Neural networks made easy (Part 17): Dimensionality reduction
18. Neural networks made easy (Part 18): Association rules
19. Neural networks made easy (Part 19): Association rules using MQL5
20. Neural networks made easy (Part 20): Autoencoders
21. Neural networks made easy (Part 21): Variational autoencoders (VAE)
22. Neural networks made easy (Part 22): Unsupervised learning of recurrent models .....
23. Neural networks made easy (Part 23): Building a tool for Transfer Learning
24. Neural networks made easy (Part 24): Improving the tool for Transfer Learning
25. Neural networks made easy (Part 25): Practicing Transfer Learning
26. Neural networks made easy (Part 26): Reinforcement Learning
27. Neural networks made easy (Part 27): Deep Q-Learning (DQN)
28. Neural networks made easy (Part 28): Policy gradient algorithm

Lawyer turned private equity titan, David Rubenstein, shares wisdom from his interviews with the world's greatest investors.

The book:
How to Invest: Masters on the Craft
by David M. Rubenstein

more...

We continue to explore reinforcement learning methods. In previous articles we discussed methods for approximating the Q-learning Reward function and the policy gradient function learning. Each method has its own advantages and disadvantages. It would be great to use the maximum of their advantages when building and training models. When trying to find methods minimizing the shortcomings of the algorithms used, we often try to build certain conglomerates from various known algorithms and methods. In this article, we will talk about a way of combining the above two algorithms into a single model training method, which is called Advantage Actor-Critic algorithm).

more...

---------------------

1. Neural networks made easy
2. Neural networks made easy (Part 2): Network training and testing
3. Neural networks made easy (Part 3): Convolutional networks
4. Neural networks made easy (Part 4): Recurrent networks
5. Neural networks made easy (Part 5): Multithreaded calculations in OpenCL
6. Neural networks made easy (Part 6): Experimenting with the neural network learning rate
7. Neural networks made easy (Part 7): Adaptive optimization methods
8. Neural networks made easy (Part 8): Attention mechanisms
9. Neural networks made easy (Part 9): Documenting the work
10. Neural networks made easy (Part 10): Multi-Head Attention
11. Neural networks made easy (Part 11): A take on GPT
12. Neural networks made easy (Part 12): Dropout
13. Neural networks made easy (Part 13): Batch Normalization
14. Neural networks made easy (Part 14): Data clustering
15. Neural networks made easy (Part 15): Data clustering using MQL5
16. Neural networks made easy (Part 16): Practical use of clustering
17. Neural networks made easy (Part 17): Dimensionality reduction
18. Neural networks made easy (Part 18): Association rules
19. Neural networks made easy (Part 19): Association rules using MQL5
20. Neural networks made easy (Part 20): Autoencoders
21. Neural networks made easy (Part 21): Variational autoencoders (VAE)
22. Neural networks made easy (Part 22): Unsupervised learning of recurrent models .....
23. Neural networks made easy (Part 23): Building a tool for Transfer Learning
24. Neural networks made easy (Part 24): Improving the tool for Transfer Learning
25. Neural networks made easy (Part 25): Practicing Transfer Learning
26. Neural networks made easy (Part 26): Reinforcement Learning
27. Neural networks made easy (Part 27): Deep Q-Learning (DQN)
28. Neural networks made easy (Part 28): Policy gradient algorithm
29. Neural networks made easy (Part 29): Advantage Actor-Critic algorithm

We continue to study model training algorithms. All the previously considered algorithms used an analytical method for determining the direction and strength of changes in model parameters during the learning process. While the methods to solve these problems turn to be inefficient. In such cases, we resort to evolutionary optimization methods.

more...

---------------------

1. Neural networks made easy
2. Neural networks made easy (Part 2): Network training and testing
3. Neural networks made easy (Part 3): Convolutional networks
4. Neural networks made easy (Part 4): Recurrent networks
5. Neural networks made easy (Part 5): Multithreaded calculations in OpenCL
6. Neural networks made easy (Part 6): Experimenting with the neural network learning rate
7. Neural networks made easy (Part 7): Adaptive optimization methods
8. Neural networks made easy (Part 8): Attention mechanisms
9. Neural networks made easy (Part 9): Documenting the work
10. Neural networks made easy (Part 10): Multi-Head Attention
11. Neural networks made easy (Part 11): A take on GPT
12. Neural networks made easy (Part 12): Dropout
13. Neural networks made easy (Part 13): Batch Normalization
14. Neural networks made easy (Part 14): Data clustering
15. Neural networks made easy (Part 15): Data clustering using MQL5
16. Neural networks made easy (Part 16): Practical use of clustering
17. Neural networks made easy (Part 17): Dimensionality reduction
18. Neural networks made easy (Part 18): Association rules
19. Neural networks made easy (Part 19): Association rules using MQL5
20. Neural networks made easy (Part 20): Autoencoders
21. Neural networks made easy (Part 21): Variational autoencoders (VAE)
22. Neural networks made easy (Part 22): Unsupervised learning of recurrent models .....
23. Neural networks made easy (Part 23): Building a tool for Transfer Learning
24. Neural networks made easy (Part 24): Improving the tool for Transfer Learning
25. Neural networks made easy (Part 25): Practicing Transfer Learning
26. Neural networks made easy (Part 26): Reinforcement Learning
27. Neural networks made easy (Part 27): Deep Q-Learning (DQN)
28. Neural networks made easy (Part 28): Policy gradient algorithm
29. Neural networks made easy (Part 29): Advantage Actor-Critic algorithm
30. Neural networks made easy (Part 30): Genetic algorithms

Advances in Financial Machine Learning
by Marcos López de Prado

Machine learning (ML) is changing virtually every aspect of our lives. Today, ML algorithms accomplish tasks that – until recently – only expert humans could perform. And finance is ripe for disruptive innovations that will transform how the following generations understand money and invest.
In the book, readers will learn how to:

• Structure big data in a way that is amenable to ML algorithms
• Conduct research with ML algorithms on big data
• Use supercomputing methods and back test their discoveries while avoiding false positives

Quantitative Technical Analysis: An integrated approach to trading system development and trading management
by Dr Howard B Bandy

I have 4 of Howard's books on my bookshelf, 'Quantitative Technical Analysis' being the latest addition. This book is rich with instruction, and where Howard's does things differently than I do, his concise explanations and logical descriptions always challenge my thinking. For example, in the book Howard discusses Cross Validation, something I have never attempted, but will now. And that is just one example, as Howard covers a lot of ground in this book. Thank you Howard, for once again significantly adding to the body of trading system knowledge.
- Kevin Davey, full time trader, author "Building Winning Algorithmic Trading Systems"

The Mental Game of Trading by Jared Tendler

A step-by-step system for mastering trading psychology.Think about your most costly and recurring trading mistakes. Chances are that they’re related to common errors, such as chasing price, cutting winners short, forcing mediocre trades, and overtrading. You’ve likely tried to fix these errors by improving your technical skills, and yet they persist. That’s because the real source of these mistakes is not technical—they actually stem from greed, fear, anger, or problems with confidence and discipline.If you are like most traders, you probably overlook or misunderstand mental and emotional obstacles. Or worse, you might think you know how to manage them, but you don’t, and end up losing control at the worst possible time. You’re leaving too much money on the table, which will either prevent you from being profitable or realizing your potential.While many trading psychology books offer sound advice, they don’t show you how to do the necessary work. That’s why you haven’t solved the problems hurting your performance. With straight talk and practical solutions, Jared Tendler brings a new voice to trading psychology. In The Mental Game of Trading, he busts myths about emotions, greed, and discipline, and shows you how to look past the obvious to identify the real reasons you’re struggling.This book is different from anything else on the market. You’ll get a step-by-step system for discovering the cause of your problems and eliminating them once and for all. And through real stories of traders from around the world who have successfully used Tendler’s system, you’ll learn how to tackle your problems, improve your day-to-day performance, and increase your profits.Whether you’re an independent or institutional trader, and regardless of whether you trade equities, forex, or cryptocurrencies, you can use this system to improve your decision-making and execution. Finally, you have a way to reach your potential as a trader. Now’s the time to make it happen.

The Little Book That Still Beats the Market by Joel Greenblatt

In 2005, Joel Greenblatt published a book that is already considered one of the classics of finance literature. In The Little Book That Beats the Market―a New York Times bestseller with 300,000 copies in print―Greenblatt explained how investors can outperform the popular market averages by simply and systematically applying a formula that seeks out good businesses when they are available at bargain prices. Now, with a new Introduction and Afterword for 2010, The Little Book That Still Beats the Market updates and expands upon the research findings from the original book. Included are data and analysis covering the recent financial crisis and model performance through the end of 2009.

In a straightforward and accessible style, the book explores the basic principles of successful stock market investing and then reveals the author's time-tested formula that makes buying above-average companies at below-average prices automatic. Though the formula has been extensively tested and is a breakthrough in the academic and professional world, Greenblatt explains it using sixth-grade math, plain language, and humor. He shows how to use his method to beat both the market and professional managers by a wide margin. You'll also learn why success eludes almost all individual and professional investors, and why the formula will continue to work even after everyone "knows" it.

While the formula may be simple, understanding why the formula works is the true key to success for investors. The book will take readers on a step-by-step journey so that they can learn the principles of value investing in a way that will provide them with a long-term strategy that they can understand and stick with through both good and bad periods for the stock market.

As the Wall Street Journal stated about the original edition, "Mr. Greenblatt says his goal was to provide advice that, while sophisticated, could be understood and followed by his five children, ages six to fifteen. They are in luck. His Little Book is one of the best, clearest guides to value investing out there."

We continue studying distributed Q-learning algorithms. Earlier we have already considered two algorithms. In the first one [4], our model learned the probabilities of receiving a reward in a given range of values. In the second algorithm [5], we used a different approach to solving the problem. We trained the model to predict the reward level with a given probability.

more...

Deep Learning and Scientific Computing with R torch (Chapman & Hall/CRC The R Series)
by Sigrid Keydana

torch is an R port of PyTorch, one of the two most-employed deep learning frameworks in industry and research. It is also an excellent tool to use in scientific computations. It is written entirely in R and C/C++.
Though still "young" as a project, R torch already has a vibrant community of users and developers. Experience shows that torch users come from a broad range of different backgrounds. This book aims to be useful to (almost) everyone.