APPLIED DEEP LEARNING WITH PYTORCH

This course provides an understanding of automating software testing using program analysis with the goal of intelligently and algorithmically creating tests. The course covers search-based test generation, combinatorial and random testing while highlighting the challenges associated with the use of automatic test generation. You will learn: Understand algorithmic test generation techniques and their use in developer testing and continuous integration. Understand how to automatically generate test cases with assertions. Have a working knowledge and experience in static and dynamic generation of tests. Have an overview knowledge in search-based testing and the use of machine learning for test generation.

This course teaches you how to build convolutional neural networks (CNN). You will learn how to design intelligent systems using deep learning for classification, annotation, and object recognition. It includes three modules: Image processing: Introduction of industrial imaging through big data and fundamentals of image processing techniques Deep learning with convolutional neural network: Overview of neural network as classifiers, introduction of convolutional neural network and Deep learning architecture. Deep learning tools: Implementation of Deep learning for Image classification and object recognition, e.g. using Keras.

The rapid development of digital technologies and advances in communications have led to gigantic amounts of data with complex structures called ‘Big data’ being produced every day at exponential growth. The aim of this course is to give the student insights in fundamental concepts of machine learning with big data as well as recent research trends in the domain. The student will learn about problems and industrial challenges through domain-based case studies. Furthermore, the student will learn to use tools to develop systems using machine-learning algorithms in big data.

The course aims to give insights in fundamental concepts of machine learning for predictive analytics to provide actionable, i.e., better and more informed decisions in, forecasting. It covers the key concepts to extract useful information and knowledge from data sets to construct predictive modeling. The course includes three modules: Introduction: overview of Predictive data analytics and Machine learning for predictive analytics. Data exploration and visualization: presents case studies from industrial application domains and discusses key technical issues related to how we can gain insights enabling to see trends and patterns in industrial data. Predictive modeling: consists of issues in construction of predictive modeling, i.e., model data and determine Machine learning algorithms for predicative analytics and techniques for model evaluation.

The course is part of the programme MAISTR (hh.se/maistr) where participants can take the entire programme or individual courses. The course is for professionals and is held online in English. Application is open as long as there is a possibility of admission. The courses qualify for credits and are free of charge for participants who are citizens of any EU or EEA country, or Switzerland, or are permanent residents in Sweden. More information can be found at antagning.se. About the course Smart Healthcare with Applications, 4 credits Who is this course for?The course suits you with any Bachelor’s degree (equivalent of 180 Swedish credit points / ECTS credits at an accredited university) who have an interest in applying Artificial Intelligence (specifically Machine Learning) to healthcare. Leadership/management experience in health-related organization/industry OR a Bachelor degree in computer science is advantageous. What will you learn from this course?Healthcare as a sector together with other health-related sources of data (municipalities, home sensors, etc.), is now in a place and can take advantage of what data science, Artificial Intelligence (AI), and machine learning (ML) have to offer. Information-driven care has the potential to build smart solutions based on the collected health data in order to achieve a holistic fact-based picture of healthcare, from an individual to system perspective. This course aims to provide a general introduction to information-driven care, challenges, applications, and opportunities. Students will get introduced to artificial intelligence and machine learning in specific, as well as some use cases of information-driven care, and gain practice on how a real-world evidence project within information-driven care is investigated. What is the format for this course?Instruction type: The lectures, announcements, and assignments of this course will be fully online via a learning management system and presented in English. Each lecture is delivered through a video conference tool with a set of presentation slides displayed online during each class session. Online practical labs (pre-written Python notebooks) are also provided in the lectures.

The course is broken down into: Basic Bayesian concepts Selecting priors, deriving some equations Bayesian inference, Parametric model estimation Sampling based methods Sequential inference (Kalman filters, particle filters) Approximate inference, variational inference Model selection (missing data) Bayesian deep neural networks