Recent & Upcoming Talks

With just a few mental tricks, an analyst can turn most optimization problems into a form that can be solved using these algorithms.

Neural embeddings are one of the most popular techniques used in Natural Language Processing to represent word similarities. There are many variations and implementations of this concept - starting from skip-gram, through GloVe or Word2Vec. But embeddings are a much more powerful concept that can be utilized in many different areas, not only NLP. Possible applications of embeddings in recommendation engines, item similarity detection, and market basket analysis through frequent items search will be presented during this speech.

How autoencoders can be used as recommendation engines, replacing collaborative filtering or other approaches.

can we use XGBoost as a time-series forecasting tool?

How can machine learning help in managing human resources in organizations?

H2O AI is one of the most interesting out-of-the-box machine learning tools. It has plenty of algorithms implemented, those algorithms are really fast and effective and it integrates well with R and Spark. One of them is especially interesing - Gradient Boosting. Want to know more?

Machine learning algorithms are replacement for an old-fashioned advisory systems. How we can utilize them in such a way? How white-box systems are different from black-box?

Presentation of various machine learning capabilities, as a tools to make sense out of large datasets