16th Deep Learning Meetup in Vienna

• Was wir unternehmen werden
Already the 16th Vienna Deep Learning Meetup is coming up - and the interest in the topic is still growing and growing!

This time our main topic is Word Embedding, which is used in Natural Language Processing (NLP), but also in sentiment analysis or searching images by textual descriptions. The talk will be complemented by a review of Andrew Ng’s Deep Learning Specialization on Coursera from a participant's perspective.

Talk 1:

Demystifying Neural Word Embedding: Applications in Financial Sentiment Analysis, and Gender Bias Detection

Navid Rekabsaz, TU Wien & Idiap Research Institute (EPFL)

A fundamental challenge in Natural Language Processing (NLP) is the understanding and capture of the semantics beyond the surface of text, especially exploring the underlying meaning of words. Word embedding models suggest machine learning-oriented methods to approach this question, so that each word is represented by a real-valued high dimensional vector. Such vectors reflect the semantical meaning of terms and have shown compelling characteristics in various NLP tasks. The word2vec algorithm introduces an effective and efficient neural network-based approach to provide such word embeddings.

In this talk, Navid will first explain the fundamentals as well as interesting qualities of the word2vec algorithm, and continue by showing some use-cases of word2vec in scenarios such as: sentiment analysis for volatility prediction in financial markets, and gender bias detection in English Wikipedia.

Talk 2:

Review of Andrew Ng’s Deep Learning Specialization on Coursera

Christoph Bonitz, Automic Software GmbH

Andrew Ng’s recent series of courses starts with the very basics of neural networks and goes all the way to applying the current state of the art in CNNs and RNNs. In this brief talk, Christoph shares his personal experience as a participant, reflecting on which of the courses may be of interest to which audience.

Hot Topics:

As usual, we will complement the talks by a presentation of latest research in Deep Learning and interesting new applications.

We hope you like the topics of our 16th Meetup and are looking forward to seeing you on 27th of February at A1 Telekom. Thanks to A1 for providing the venue and catering.

Best regards,
Tom Lidy, Jan Schlüter and Alexander Schindler

Please note that the planned discussion panel on "Ethics and Bias in AI" has been postponed to May.