Machine Learning Talks @ IMC - Independently organized event

This is an independently organized event by IMC - supported by the Machine Learning Netherlands Meetup.

IMC is a technology-driven trading firm. Find more information about IMC here (https://www.imc.com/eu/).

Schedule

06:00 PM – Welcome with food and drinks

07:00 PM – Introduction to IMC

07:05 PM – First talk by Thomas Kipf (PhD student in Machine Learning at University of Amsterdam)

07:50 PM – Break

08:00 PM – Second talk by Edwin de Jong (Machine Learning Researcher at IMC)

08:30 PM – Network drinks

09:30 PM – The end

If you would like to socialize a later than 9:30 PM, IMC recommends going to WTCafé De Blauwe Engel. This café is located in the WTC, on your way out to train station Amsterdam Zuid.

Don’t worry if you cannot make it there by 6:00 PM but try to be present just before the first talk starts at 7:00 PM.

First talk by Thomas Kipf (PhD student in Machine Learning at University of Amsterdam)

Thomas Kipf is a second-year PhD student in Deep Learning for Network Analysis at the University of Amsterdam, supervised by Prof. Max Welling. His main area of interest is large-scale inference for structured data and semi-supervised learning. He further explores topics in reasoning and multi-agent reinforcement learning. His formal background is in Physics (M.Sc. hons. 2016, B.Sc. 2014 at FAU). During his studies, he has had exposure to a number of fields and—after a short interlude in Neuroscience-related research at the Max Planck Institute for Brain Research—eventually developed a deep interest in Machine Learning.

Deep Learning on Graphs with Graph Convolutional Networks

Deep learning has recently enabled breakthroughs in the fields of computer vision and natural language processing. Little attention, however, has been devoted to the generalization of deep neural network-based models to datasets that come in the form of graphs or networks (e.g. social networks, knowledge graphs or protein-interaction networks). In this talk, Thomas will introduce graph convolutional networks (GCNs), a recent class of models that operate directly on graphs. GCNs compare favorably against established graph-based methods and provide a new approach for problems such as link prediction or node classification.

Second talk by Edwin de Jong (Machine Learning Researcher at IMC)

Edwin de Jong has a lifelong interest in exploring and creating new machine learning and artificial intelligence technology. After receiving his PhD from the VUB AI Lab and doing a postdoc at Prof. Jordan Pollack’s lab at Brandeis University (Boston, MA), he continued in machine learning research and published over 60 scientific articles (https://scholar.google.com/citations?user=l9w80gcAAAAJ&hl=en). To ensure this new technology also finds its way to industry, in 2005 Edwin co-founded Adapticon, one of the earliest Deep Learning startups worldwide. Next, at Quintiq, he led the development of Quintiq’s Demand Planner product, and later became responsible for all Predictive Analytics technology. As of April 1st 2017, Edwin will be joining IMC (https://www.imc.com/us/about-us#what-we-do) as a Machine Learning Researcher.

Incremental Sequence Learning

In this talk Edwin de Jong will touch on some recent examples of the impressive progress of Machine Learning research, and share results of his own research into Incremental Sequence Learning (https://edwin-de-jong.github.io/blog/isl/incremental-sequence-learning.html) which was presented at the NIPS 2016 CLDN (https://sites.google.com/site/cldlnips2016/) workshop.

Please note: IMC will be able to host approx. 50 attendees with a small chance of not having enough seats if more people choose to attend. A few people may have to stand/lean during the talk.