Recent Posts by David Gruzman

HW selection benchmark for deep learning

Our company is doing active research in the area of deep learning. To be a bit more specific - we are building very wide embedding layers. We need to be able to perform an interactive research and we need to be able to train a lot of networks on-demand. The number of weights in our…
Read More

Power of Comparison

Our technology is based on the comparison of data segments. We are capable of detecting a similarity degree of the data from different servers, from different buildings, etc. Our main objective is anomaly detection. In other words, we are looking for segments that are unique and do not resemble any of the previous events. However,…
Read More

Participation in Kenshoo data hackathon

Some time ago we got an offer from Shlomi Hassan to join Kenshoo hackathon (http://kenshoo.com/data-hackathon/). The idea of hackathon was to try to perform a kind of user value prediction based on history of events. It has to be said that predictive models are not exactly our specialisation - we are working on anomaly detection…
Read More

What is an interesting segmentation?

Let's consider a situation when you have a new data set. You bought it, got it from some external source of yours, etc. The important fact is that this is a new dataset, and you are not familiar with it. In fact, you barely understand the semantics of the columns. From where would you begin?…
Read More

Recent Comments by David Gruzman

No comments by David Gruzman yet.