Okay, let's continue from where we stopped the last time and talk a bit about the high level objectives that you will have in this specialization. Your first objective should be to learn the core concepts of machine learning such as types of machine learning, notions of trading count, test sets, all featuring categorization, and so on, and so forth. Second, you want to understand how specific machine learning algorithms such as logistic regression, support vector machines, decision trees, and solve work for financial problems. Further you want to learn more about neural networks maybe because you heard a lot about deep learning, and you want to find out what neural networks and deep learning are about and how they're used in finance. Next you want to get that practical experience with using Python Machine Learning libraries on financial problems including in particular, search libraries as scikit-learn for classical machinery algorithms, and library such as TensorFlow for neural models. Finally, you want to know how you would be able to expand your knowledge and learn more after you will complete this specialization. All of the above will be our objectives here and we will pursue them both in sequence by progressing through courses in this specialization. Let me explain what we will do in each one of them. Our first course is an introductory course that I called a Guided Tour of Machine Learning in Finance. This course will help you to grasp basic concepts of machine learning and understand how in general machine learning in finance differs from machine learning in the tech industry. You will learn some simple machine learning algorithms that we will use to analyze stock returns, company fundamentals, bank reports, and other financial data. All with examples that present problems of practical interest. I will show you Jupyter Notebooks that implement these simple algorithms and then you will extend them using different data in different algorithms. In this first course, we will be mostly using scikit-learn, and a few other packages to implement our machine learning solutions. Simultaneously, we will gently introduce TensorFlow, a very powerful Python package for machine learning, open source by Google about two years ago. Though our main interest in TensorFlow is due to its power of functionality for neural networks. We will start with most simple examples of using TensorFlow such as linear regression or logistic regression. What comes to using neural networks themselves, we will get a glimpse of how they work in the first course and build our simple neural network for analysis of company earnings and bank statements. During this course, we will have weekly home assignments that will include programming assignments and optional reading. The final course projects will be to apply several supervised learning algorithms to bank reports data. Don't worry now if you don't know what the term supervised learning means, we will learn it soon. Your final score for the course will be computed as a cumulative score of your weekly homework assignments and your final course project. Once you completed the first course, the four work courses cover the main topics of machine learning paradigms. For now, let me just name them for you. They are called supervised learning, unsupervised learning, and reinforcement learning. Respectively, if each one of the follow up courses will extend our first course in both the depth and the implications considered. In our second course, we will go into more details of supervised and unsupervised learning, can get familiar with many key algorithms that belong in these types of machine learning. We will be talking about various types of regression and classification problems and see how they can be solved using methods such as logistic regression, support vector machines, or various types of decision trees. Then, we will turn to unsupervised learning and talk about methods of data analysis and data visualization such as principal component analysis and its extensions, clustering methods, and so on. We will also look into neural networks and see how they can be used to address the same types of tasks of supervised and unsupervised learning. On a financial side, we will be looking at many classical problems of finance such as market and regime forecasts, predictions of stock returns, portfolio optimization, etc., and show how machine learning is applied for these tasks. If you're only interested in financial implications of supervised learning or unsupervised learning, the first two courses should give you strong fundamentals in understanding core concepts and main approaches in these fields. So, that you would be able to go on and use them in real life after completing these two courses. My personal view though is that the most fascinating implications of machine learning in finance come from reinforcement learning. We will touch upon reinforcement learning in our first two courses, and we will fully devote our third course in this specialization to reinforcement learning, and its financial implications. In this course, we will talk about using of reinforcement learning to solve such classical problems of finance as portfolio optimization, optimal trading, and option pricing and risk management. This third course is a bit more technically advanced than the first two courses and we'll have a bit more math in it, but not by much. The main difference of this course from other courses on reinforcement learning, is that we will spend exactly zero time talking about such traditional examples of reinforcement learning as Tic-Tac-Toe or robotics. Instead we'll jump to financial problems right from the start and learn classical methods of reinforcement learning such as the famous Q- learning using financial problems. This will be a hands-on course where you will be implementing such algorithms and look how they work in practice. In your course project for this course, you will explore simple model for market dynamics that is obtained using reinforcement learning. After that, our final course, we'll go deeper into topics of reinforcement learning that we address in this third course. Through the fourth course, we will focus on more advanced topics, but it will be less technical than the third one, and will be structured as an overview of selected topics. The first class of topics will focus on fundamental concepts of finance such as market equilibrium, no arbitrage, predictability, and so on, and how reinforcement learning methods offer new views for such concepts. We will also talk more about market modeling cues in reinforcement learning. The second part of this last course will overview applications of reinforcement learning to high frequency trading, credit risk in peer-to-peer lending, and cryptocurrencies trading. Finally, the last thing I need to discuss before we move on with the first course in this specialization are prerequisites to this course, textbooks, and course resources. Let's talk about it in the next video.