TALKS AND SEMINARS

The first time you say something, it’s heard; the second time, it’s recognized; the third time, it’s learned

John Maxwell

SPIE Photonics West Conference, 2nd-9th February 2019, S. Francisco, California, USA

Prof. Dr. Venturini will present the results of our last work at the SPIE Photonics Conference in San Francisco, California, USA.

New approach for luminescence sensing based on machine learning
Paper 10937-14
Time: 4:10 PM – 4:35 PM
Author(s): Francesca Venturini, Michael Baumgartner, Zürcher Hochschule für Angewandte Wissenschaften (Switzerland); Umberto Michelucci, TOELT LLC (Switzerland)

Abstract

Luminescence sensors are based on the determination of emitted intensity or decay time when a luminophore is in contact with its environment. Typically, since the absolute values of the measured quantities depend on the specifc sensing element and scheme used, a sensor needs an analytical model to describe the dependence of the quantity to be determined, for example the oxygen concentration concentration, from sensed quantity, for example the decay time. Additionally, since the details of this dependence are device specifc, a sensor needs to be calibrated at known reference conditions. This work explores an entirely new artifcial intelligence approach and demonstrates the feasibility of oxygen sensing through machine learning. After training the neural network on synthetic data, it was tested on measured data to verify the prediction of the model. The results show a mean deviation of the predicted from the measured concentration of 0.5 % air, which is comparable to many commercial and low-cost sensors. The accuracy of the model predictions is limited by the ability of the generated data to describe the measured data, opening up future possibilities for signifcant improvement by performing the training on experimental data. In this work the approach is tested at different temperatures, showing its applicability in the entire range relevant for biological applications.

 

WORLD WEB FORUM, 17th-18th January 2019, Zürich, Switzerland

BIG DATA & MACHINE LEARNING – PERLS & PITFALLS

I gave a keynote a the World Web Forum (https://worldwebforum.com/) in Zürich on the 17th-18th of January. I talked about what is AI and Machine Learning, and gave an introduction to everyone and discuss what the future will hold. I was in the real estate stream but my talk is quite general.

Workshop, KTI Project, Collaboration with University St. Gallen and ZHAW, Nov. 2018, St. Gallen, Switzerland

Die unsichtbare Herausforderung der KI: die Integration von Text Mining Algorithmen in eine reale produktive Umgebung

In November I gave a talk at a workshop in German with the title “Die unsichtbare Herausforderung der KI: die Integration von Text Mining Algorithmen in eine reale produktive Umgebung” dealing on how to integrate text mining algorithm in a productive environment in collaboration with two universities, the University of St. Gallen and the ZHAW.

TDWI Conference, 5th-6th November 2018, Zürich, Switzerland

The future of Analytics and Innovation – an unknown land

In November I gave a talk at the TDWI conference in Zürich. Big corporations are typically very slow in innovating and usually don‘t manage to do any real research. Why is so difficult for them? The Analytics field seems predestined for innovation and research, but still very few companies are really able to use their data at their full potential. Many don‘t even really know what data they have, or how to use it and profit from it. Is it possible to change all that? It is not only possible, but a necessary step for big corporations to survive in the next years. Innovation and research are not solutions to isolated problems, but the way we will need to transform the way we work in the future to be successful in any market.

Erfahrungsbericht: Wenn Forschung auf produktive Umgebung trifft, Zürich, November 2018

Nach der Entwicklung von Text-Mining-Algorithmen zur Erkennung von Duplikaten beim Testen von Tickets stand Helsana vor einer echten Herausforderung. Wie integriert man die Technologie in eine heterogene Umgebung? Wie kann Künstliche Intelligenz mit vielen anderen Technologien gut funktionieren?
In diesem Vortrag  werde ich einen Überblick darüber geben, welche Herangehensweise sie ergriffen haben, um die technischen und organisatorischen Herausforderungen in einem forschungsbasierten Projekt zu bewältigen, und wie die Forschungszusammenarbeit mit Universitäten und Fachhochschulen uns geholfen hat, erfolgreich zu sein.

  • Wie unterscheidet sich ein forschungsbasiertes Projekt von einem klassischen
  • Wie Forschungskooperationen Unternehmen in KI-Projekten helfen können
  • Was ist die größte Herausforderung? Die Technologie oder die Unternehmenskultur?

Re-imagining education: the solution to embracing Artificial Intelligence in our society, September 2018

I gave a talk at the Swisscognitive event in January 2018 in Zürich with the title Re-imagining education: the solution to embracing Artificial Intelligence in our society.

Abstract: 

One of the biggest challenges in all industries is how to make AI a fundamental part of a company strategy and culture. What are the use cases? How to be successful? How to profit from it? Everyone seems to be struggling in answering those questions. We argue that the most fundamental reason is that the present education system is training the very skills that we need to let go in the future. We need to train fundamental new skills like creativity, critical thinking, research design in a much earlier phase of the curriculum. We need to disrupt the actual education system to make AI integration in all industries common place. Companies need to participate actively in the shaping of the curriculum of tomorrow and cannot rely anymore on the academic world alone.

In this talk we will look at how we need to change education at different levels, to enable companies to profit from AI, and to make it a fundamental component of their strategy and their culture. We will discuss what skills are missing in an AI and data science centred Curriculum today and what we are doing to change all that.

Image source: swisscognitive.ch

Defining the undefinable concept of innovation: impossible?
SAS Forum Switzerland, Invited Talk, Zürich, Switzerland, June 2018

Today everyone is talking about innovation. Every company, startup or university has innovation experts, departments or teams. The problem is that everyone has a different understanding of innovation. Every industry (and even departments in the same industry) has different needs and requirements for what relates to innovation. Giving a unique definition of innovation seems impossible, and nobody has yet managed this seemingly impossible task. I split the concept of innovation in two parts: finding and formulating problems and then solving them. I argue that the problems that innovation tries to solve are by definition not solvable by existing methods, and therefore require new paradigms to be solved. I will analyze the type of problems that innovation is trying to solve and show how they differ from what companies have faced in the last ten years. If we understand the types of problems we need to solve we understand what it means to innovate and what we need to do it. I will give an interpretation of innovation that is not industry related and one that I believe should drive the efforts of companies to change their culture and way of working in the next years. Many examples of famous innovations will support and make my arguments clear. In particular I look at how SAS is a tool that can support innovation efforts of Companies in the area of Analytics and new technologies.

Können Dinosaurier Ballettschuhe tragen?
Invited Talk, SAS Forum, Bonn, Germany, May 2018

Today everyone is talking about innovation. Every company, startup or university has innovation experts, departments or teams. The problem is that everyone has a different understanding of innovation. Every industry (and even departments in the same industry) has different needs and requirements for what relates to innovation. Giving a unique definition of innovation seems impossible, and nobody has yet managed this seemingly impossible task. I split the concept of innovation in two parts: finding and formulating problems and then solving them. I argue that the problems that innovation tries to solve are by definition not solvable by existing methods, and therefore require new paradigms to be solved. I will analyze the type of problems that innovation is trying to solve and show how they differ from what companies have faced in the last ten years. If we understand the types of problems we need to solve we understand what it means to innovate and what we need to do it. I will give an interpretation of innovation that is not industry related and one that I believe should drive the efforts of companies to change their culture and way of working in the next years. Many examples of famous innovations will support and make my arguments clear. In particular I look at how SAS is a tool that can support innovation efforts of Companies in the area of Analytics and new technologies.