Book Image

Mastering Machine Learning Algorithms. - Second Edition

By : Giuseppe Bonaccorso
Book Image

Mastering Machine Learning Algorithms. - Second Edition

By: Giuseppe Bonaccorso

Overview of this book

Mastering Machine Learning Algorithms, Second Edition helps you harness the real power of machine learning algorithms in order to implement smarter ways of meeting today's overwhelming data needs. This newly updated and revised guide will help you master algorithms used widely in semi-supervised learning, reinforcement learning, supervised learning, and unsupervised learning domains. You will use all the modern libraries from the Python ecosystem – including NumPy and Keras – to extract features from varied complexities of data. Ranging from Bayesian models to the Markov chain Monte Carlo algorithm to Hidden Markov models, this machine learning book teaches you how to extract features from your dataset, perform complex dimensionality reduction, and train supervised and semi-supervised models by making use of Python-based libraries such as scikit-learn. You will also discover practical applications for complex techniques such as maximum likelihood estimation, Hebbian learning, and ensemble learning, and how to use TensorFlow 2.x to train effective deep neural networks. By the end of this book, you will be ready to implement and solve end-to-end machine learning problems and use case scenarios.
Table of Contents (28 chapters)
26
Other Books You May Enjoy
27
Index

Direct policy search through policy gradient

The last method we're going to discuss doesn't employ a proxy to find the optimal policy, but rather looks for it directly. In this context, we always assume we're working with a stochastic environment where the transitions are not fully deterministic. For example, a robot can assign a high probability to a transition but, in order to increase the robustness, it has also to include a noise term that can lead the transition to a different state. Therefore, we need to include a transition probability .

Given this element, we can evaluate the overall probability of an entire sequence (that normally ends in a final state): Sk = (s1, s2, …, sk). To do so, we need to define a parameterized policy ; therefore, the probability of Sk can be expressed as a conditional one: . The full expression requires us to explicitly separate the state transition component from the policy and to introduce the actions (which were implicit...