You don’t have to sign up for talks.
L2F – Learn To Forecast
Wednesday 27.02.2019 17:30 HG D 7.2
Machine Learning through the lens of Topological Data Analysis
During this talk we would like to present the scientific mission behind our research program in applied topology as well as our past, current and future projects.
II-VI Laser Enterprise
Thursday 28.02.2019 17:30 HG D 7.2
How many lasers does man need? Production and applications of AlGaAs lasers
While at the beginning of the 80s the CD player, the first consumer application for lasers, found its way into almost every household under great attention, today we almost constantly use laser diodes, often without noticing it.
In this lecture the physical basics of AlGaAs semiconductor lasers will be discussed first. Subsequently, the manufacturing process of the laser diodes is briefly described.
The third part discusses some of the above applications in more detail to give an impression of the variety of applications and their importance for our lives today.
D | ONE
Friday 01.03.2019 10:15 HG E 3
Real-time automated insurance claim processing: Using NLP methods to learn sparse spaces
Claim processing in insurance is at the core of its operations and business. Even in the case when claims are processed entirely electronically it is rather hard to assess all of the new claims automatically. The industry standard is to use Rule Engines with a large set of expert-written rules. The limitation to this approach is increasing complexity of rules and the related maintenance. Additionally these systems fail to classify a significant portion of incoming claims.
In this talk you will see the automation of the processing for complex claims using machine learning methods: the journey with our client starting with an initial proof of concept, through the definition of the business case, and the implementation of the machine learning algorithm, all the way to the deployment and the operationalisation of the new system.
A particularly focus are the used state of the art NLP methods, namely word2vec, to define a similarity measure and learn a sparse space for 100’000 features. Additionally you will see how we built an intuitive validation framework, thus building trust in the approach.
Friday 01.03.2019 11:15 HG E 3
A Head Full of Answers – Introducing a Modular Cognitive AI Platform at Credit Suisse
provide insight into how we moved from vision to delivery, changing the business model and transforming the organization by implementing and pushing for adoption of a cognitive AI platform
- outline the start from the story of devising a strategy to digitalize a classic, incumbent support function to reaching the tipping point of making a platform case across business functions
- share our learnings along this experimental journey: selling a vision, not yet facts; working in a “fluid” governance; integrate whilst staying open for innovation at the same time; transform doubters into believers across the organization;
- show a concrete example of full agile delivery for the right focus on user journey versus MVP (minimum viable product)
Friday 01.03.2019 13:15 HG E 3
Why a train is a blockchain too
We want to give you an overview what we have achieved with Blockchain-Technology at Swiss Federal Railway. In addition, we will also give you insights into a showcase on how we secure one of the greatest buildings in the history of mankind with the greatest technology which ever existed.
Friday 01.03.2019 14:15 HG E 3
Aktuare bei Zurich
Einblick in den Arbeitsalltag als Aktuar im Aktuariat der Zürich Leben Schweiz anhand von zwei Praxisbeispielen: R für Risiko und Rückversicherung, ein Praxisbeispiel. Einsatz von R Shiny für die Berechnung von StopLoss Offerten. Bestimmung von Best Estimate Annahmen für Sterblichkeit.