Two more talks are coming up for this month: one industry speaker and one academic speaker. The industry talk will be given by Damien Lefortier from Criteo (http://www.criteo.com) (see the abstract below). Prof.dr. Marcel Worring (https://staff.fnwi.uva.nl/m.worring/prof.html) (University of Amsterdam) will give the academic talk. He will talk about multimedia search/representation/visualization.
Damien Lefortier(Criteo (http://www.criteo.com))-- Machine Learning for Display Advertising @ Scale
The field of online advertising represents an exciting opportunity for machine learning engineers and scientists: large volumes of data, fast-pace environment enabling quick iterations of research ideas, efficient performance-based marketplace allowing to precisely measure ad effectiveness at a worldwide scale. Industrial players have leveraged the recent advances in large-scale computation to build facilities capable of hosting massive amounts of data and CPU-hungry algorithms. For all these reasons, the last decade has seen tremendous progress in the application of machine learning to online marketing.
In this talk, we will briefly introduce the online advertising marketplace, its stakeholders and the key performance metrics. We will then present the algorithms we have developed at Criteo for bidding in real-time auctions and product recommendation at scale and explain how we evaluate them both offline and online. We will describe the infrastructure for large-scale data processing that these algorithms rely upon. Finally, we will conclude with future areas of research and open the floor for a panel discussion with participants.
Damien Lefortier is a Senior Software Engineer and Tech Lead in the Prediction Machine Learning team at Criteo where he has been actively involved in the development of Criteo's large scale distributed machine learning library as well as in improving Criteo's predictive algorithms for ad targeting. At the same time, he started his PhD in information retrieval at the University of Amsterdam under the supervision of Maarten de Rijke. His research work has been published at the top tier conferences, such as WWW and CIKM.
Prof.dr. Marcel Worring (https://staff.fnwi.uva.nl/m.worring/prof.html) (University of Amsterdam)--Image Search Beyond the Ranked List
The search engine paradigm in which the result is always a single ranked list is a very powerful mechanism. Yet
for explorative tasks it has many limitations. We argue that for image set exploration iterative categorization by human and machine is an appropriate model and to be effective should be supported by machine learning, summarization, and visualization. We have brought all of these together in the multimedia pivot table concept. We will elaborate on the design, underlying algorithms, its use in a number of application scenarios, and consider how to evaluate such complex explorative systems.
Marcel Worring is an associate professor in the Informatics Institute and a full professor in the Amsterdam Business School, both at the University of Amsterdam. He is associate director of Amsterdam Data Science a collaboration of CWI, HvA, VU and UVA on the broad topic of data science. He (co-)authored more than 200 scientific publications focusing on image/video retrieval and more recently visual analytics. He is general co-chair of the upcoming ACM Multimedia in Amsterdam, the leading conference on the topic.