Michael Cochez, academic lab manager of the discovery lab was invited to give a keynote at the 3rd Workshop DL4LD: Addressing Deep Learning, Relation Extraction, and Linguistic Data

In his talk, Michael talked about several core themes of the lab: machine learning, knowledge representation, and scientific corpora.

abstract

In the past decade, deep learning has taken a prominent role in the research landscape and some applications, often related to perception or generation, have become mainstream. Already before that, we have seen the uptake of technologies to represent complex interconnected information in the form of a knowledge graph. In this talk, we will explore several ways in which this world of deep learning and the world of rich semantic information interact with language. We will do this with the help of several research challenges, like query answering, incorporating labels into link prediction and inductive predictions, and scaling neural network training.