November Meetup @ IronScales

We would like to thank IronScales for hosting us, we have a great event planned for you:

18:00-18:30 Gathering & Mingling
18:30-18:45 A word from our sponsor
18:45-19:15 Data enrichment and contextual data pipelines for real-time prediction systems (Mordechai / IRONSCALES)
19:15-19:30 Break
19:30-20:00 What should I buy next? How to leverage word embeddings to build an efficient recommender system. (Aleksander Molak)
20:00-20:30 Audio psychoacoustics and speech quality measurements(Orian Sharoni / Up.AI)

====================
Data enrichment and contextual data pipelines for real-time prediction systems (Mordechai / IRONSCALES)

Contextual data is a critical element for a robust ml system yet building real-time data pipelines is very challenging. In this talk, we describe Ironscales' data systems, how we scaled to loads of 50k predictions per second, and how contextual ML models are used in ensemble with other classifiers

===============================
What should I buy next? How to leverage word embeddings to build an efficient recommender system. / Aleksander Molak

Word embeddings revolutionized natural language processing in the last decade. They were and are widely used in many applications from academia to business and powered the recent Transformer revolution. In this talk, we will look on how the power of embeddings can be leveraged to build an efficient recommender system. I will share my experiences based on research and real-life projects for some of our global customers.

============================
Audio psychoacoustics and speech quality measurements(Orian Sharoni / Up.AI)

The missing metric and how to train models without it

The gap between perceived audio quality and measured signals has been the subject of research for engineers and philosophers for the past 120 years. The talk will recap where we came from, how we train models now and where we are heading.