Wearables and Machine Learning

The course consists of lectures and practical exercises in combination of wearables and machine learning. It includes a full-framed introduction to the latest methods and their applicability to the real-life issues. The workshop is presented in cooperation with the Uplinx-Project (@German Research Center for Artificial Intelligence) and the Design Research Lab (@UdK Berlin).

Technological progress of the last years has caused major changes in design and clothing industry. The wide availability of digital components and accessories enabled their integration into clothes and wearable accessories. The affordable prices made the products accessible to a big part of the market and widely recognized under the notion of Wearables. Meanwhile, this enthusiastic response of the consumer base has led the industry to further research and development in order to fulfill the more challenging requirements which can be applied to the users' everyday life. An important aspect here is the possibility of using this technology to provide solutions for the health sector and people with disabilities, giving them better perspectives for inclusion and equal opportunities.

A very basic and important attribute of the new wearable technologies is the fact that these accessories accompany their users in a majority of their regular activities, most of them on a daily basis. The first generation of wearables offered intelligent functions with pre-programmed rules based on common knowledge and general world behavior. Nevertheless, the ability to follow and store the activities of the user gives the possibility to enhance the intelligent functions with empirical knowledge. This way, the accessories are capable of adapting themselves to the needs of the individual users and therefore being more useful by the time. These features are actively backed by state-of-the-art technologies which are associated with a broad spectrum of scientific research in the area of  informatics. These technologies include Big Data which allows an efficient storage and processing of the data gathered by the devices, as well as Machine Learning which offers the statistical background to analyze this data, extract the important patterns and transform them into useful actions for the user. Machine Learning is considered the core technology that is expected to shape the future in personalized technologies.

In this course, we are therefore focusing on this combination of Wearables and Machine Learning. We are aiming to offer the participants a full-framed knowledge base about the methods that form the current state of the art and give them the possibility to understand their application through practical exercises.

PARTICIPANTS

The course is aimed at international students from technical and design studies. The course is free of charge. Students need to apply through the website and send a short motivational statement and a CV until June 2nd. Afterwards there will be a selection of 15 participants.

BACKGROUND KNOWLEDGE AND REQUIREMENTS

The course will be adapted to the background of the participants. The class will involve interdisciplinary work in groups. This enables an exchange of knowledge and experience within the groups depending on the background of the participants. Participants with limited background knowledge will be given creative introductory exercises whereas participants with a background in programming or Arduino will have the possibility to work on more advanced topics. Participants are therefore advised to submit a CV or a statement about their background knowledge prior to the beginning of the summer school in order to assist the proper preparation of the class.


SCHEDULE

Day 1: Input and motivation.

• Presentation by the invited expert(s)
• Application of the first design methods to analyze a problem and suggest possible solutions

Day 2: Theoretical input to Machine Learning and E-Textile Technologies

• Introduction to simple Arduino circuits
• Presentation of the Fiber Space lab

Day 3: Experimenting with particular tasks through the use of advanced design methods

• Support with the programming of models
• Construction of Arduino circuits

Day 4: Interdisciplinary Quick and Dirty Prototyping with different materials and techniques

Day 5: Analysing the results of the exercises, basic feedback and a possibility for improvements.
Presentation of material and suggested tasks for further work.

Eleftherios Avramidis is a researcher specialized on Machine Learning and Artificial Intelligence. He has been a researcher at the German Research Center for Artificial Intelligence (DFKI) since 2010. He joined the Design Research Lab and the Interactive Textiles Lab of DFKI in November 2017. He has been active with the development of machine learning models that predict the user qualitative preferences on text outputs and he is a Ph.D. candidate at the University of Saarland with the topic of Quality Estimation for Machine Translation.

Friederike Fröbel is a researcher at the German Research Center for Artificial Intelligence
(DFKI) in the research group of Interactive Textiles. In January 2017 she joined the Design
Research Lab as a student assistant and pursued her MSc degree in 2018 with her thesis
entitled “Smart Fashion as Mobile User Interface”, supervised by Dr. Katharina Bredies. The
thesis involved the prototypical development of a mobile application as a user interface for a
service system in Smart Fashion.