Dear all
Manythanks for your participation at our conference. Below please find a shortrecap and an outlook for our next conferenceon September 6, 2018.
Aimof the conference
The aim of this conference wasto bring togetherEuropean academics, young researchers, students and industrial practitionersto discuss the application of Artificial Intelligence to variouspractical fields. In a broader context, we wanted to promote ?Mathematicsfor Industry? in Switzerland, as part of the European COST (Cooperation inScience and Technology) Action “Mathematics for Industry”, where members ofZHAW are in the management committee for Switzerland.
COSTis the longest-running European framework supporting transnational cooperationamong researchers, engineers and scholars across Europe.The 1stEuropean COST Conference in Switzerland on this topic was held onSeptember 15, 2016.A third installment is planned for Thursday,September 6, 2018.
Speakers
Theconference beganwith the key-note speech on “MachineLearning in Trading”byProf. BrianPeterson, from the University of Washington and DV Trading, givinginsight in the impact and extent of this emerging technology within itsapplicational frame in trading.
Following this, we had twoparallel thematic sessions.One leg focused on Financial Mathematics applicationof machine learning, whereas the other one dealt with the implications forIndustrial Mathematics.
Please find a short list of the speakersbelow, with anexcerpt attached as a summary of their presentation.
Finance
SaeedAmen,Cuemacro
Using Big Data to trade Macro Assets
“Wediscuss the types of unusual datasets which are available for trading purposes.We define the difference between unstructured and structured datasets. We alsogive a case study of using Big Data, creating macro-economic sentiment indicesfrom RavenPack news data, showing how they can be used to trade bonds and FX.”
Dr. Pascal B?hi, FintegralAG
CreditRisk Evaluation with Machine Learning Techniques
“This talk will cover different data analysis andmachine learning techniques, and analyze the automation of the variable/featureselection process for the generation of robust credit scoring models.Furthermore, benchmarking of the aforementioned techniques to well-established,industry standards will be demonstrated.”
Dr.Christopher Bruaffaerts, InCube Group
Recommender System for PersonalizedInvestment
“In this talk, we will see howCase-based recommender systems can take into account the specificities andconstraints of the financial domain to appropriately recommend investmentproposals”
Prof. Dr. Matthias Fengler,Universit?t St. Gallen
DoesSentiment drive the Market?
“We find that options variables indeed predict stock returns,yet sentiment variables, in particular, our index sentiment remains a highlyrelevant factor for individual stock returns”
Dr.Julian Lorenz, Bantleon Bank
Forecasting financial Markets usingboosted Decision Trees
Prof. Dr. Natalie Packham,Berlin School of Economics and Law
TailRisk Protection Trading Strategies
“Finally, by empirically testing for second-orderstochastic dominance, we find that risk averse investors would be willing topay a positive premium to move from a static buy-and-hold investment in the DAXfuture to the tail-risk protection strategy.”
Dr. Jochen Papenbrock, Firamis AG
Financial Networks and related AI inFinancial Services
“Inthis talk we will give an overview of the methods and present some use cases inasset management and banking. The related software platform architectures willalso be discussed.”
Henrik Stutz
PredictingTail Events for Equity Indices using Machine Learning Algorithms
“Inthis talk, results for the tail prediction of events during the closing hoursof various stock markets were analyzed. Namely the methods in question wereCART, Random Forests, Bagging and Boosting.”
Prof. Dr. Mario Wüthrich, ETH Zürich
MachineLearning applied to Non-Life Insurance Problems
“Theaim of this presentation is to bridge the gap between the actuarial communityand the machine learning community. We present actuarial problems and discussmachine learning methods.”
Industry
Dr.Juan Pablo Carbajal, EAWAG
Speeding up numerical Simulators withEmulators
“Inthis talk I will introduce the latter (fast surrogate
model)approach, describe its relation to model order reduction, comment on thegeneral-purpose vs. specific trade off in terms of efficiency of computation,and present some recent advances in emulation methodologies.”
Dr.Martin Fengler, Meteomatics
Solar and Wind Power Forecasting
“Thistalk is about introducing the different modeling approaches which involvemachine learning methods.”
“Wepresent decision trees (CART) and boosted decision trees (using AdaBoost) inthe context of forecasting financial markets, formulating the forecast problemas binary classification in a supervised learning setup.”
Prof. Dr. Joachim Giesen, Universit?t Jena
Matrixcalculus.org
“Inthis talk I will demonstrate the tool using the classroom example of logisticregression from the domain of machine learning.”
Dr. Christian J?ger, ZHAW
Energy Management in HVAC, from classicaloptimization to artificial Intelligence
“Thistalk presents results and insights from the development of a control systembased on classical optimization algorithms focusing on resource-efficiency aswell as the outline of a planned follow-up project, aimed at using machinelearning methods to adapt to inhabitants needs and preferences.”
Dr. David Mordaunt, Precision Light
4DImagining and Modelling for plastic Surgery
This talk shed light on the application ofspatial and temporal body modelling for the prediction of the results ofsurgical procedures.
Dr. Hans Rudolf Moser, UZH
Generalized Entropies in empirical Time Series:Inherent Properties and forecasting Intervals
“Atheoretically as well as practically powerful outcome of dynamical systemstheory is the possibility to determine dynamical invariants by virtue of along-term integration.”
Dr. Roxana Istrate, IBM
IncrementalTraining of Deep Convolutional Neural Networks
“We propose anincremental training method that partitions the original network intosub-networks, which are then gradually incorporated in the running networkduring the training process.”
Dr. Christian Spindler, PWC
Automated Machine Learning in transactionalData
“Thetalk gives an introduction into PwC’s proprietary solution for automatedmachine learning. By means of our Data Science Machine, we are able to find andoptimal set of algorithm, hyperparameter tuning and selected features to solvea given problem.”
Dr. Sebastiano Vascon, ECLT
HumeNash Machines: Context-Aware Models of Learning and Recognition
“In this talk I’lldescribe a framework for pattern recognition problems which is grounded in theprimacy of relational and contextual information at both the object and thecategory levels.”
Participants
The number of participants exceeded our initial estimates. Overall,we had more than 190 registrations. Regionally, we covered all of Switzerland,with the focusonthe larger Zurich area.
This iteration of the COST conferencesaw a large number of international guests and speakers, travellingtoSwitzerland from destinations such as the UK, Germany, the United States andBulgaria.
We have had a large proportion ofrepresentatives from the industrycomplemented by a smaller number of academicresearchers.