Machine Learning with Embeddings & Explaining Deep Neural Networks
Talk 1:
📢: Maria Perchyk, software engineer
🗣: Machine Learning with Embeddings
Embeddings are one of today’s main concepts in machine learning. They capture the semantics of all kinds of data, such as texts, images and videos. Image classification and sentiment analysis are only two out of many applications. In this talk, I will show you how to visualize, train and apply embeddings for text and image data.
Talk 2:
📢: Oliver Zeigermann, developer/data scientist (http://zeigermann.eu/)
:🗣: Explaining Deep Neural Networks using Surrogate Models
Deep Neural Networks are great for learning any form of decision boundaries and also can be made to generalize well. But with a lot of complexity they become black boxes and no one can understand what is going on inside of them. In this talk I will demo the idea of a decision tree as a surrogate model for such a network that gives you a dial between understandability and accuracy
#machine-learning #data-science #python #tensorflow
