How you learn big things with little data

We start Y-DATA talk series called Data Driven Products. This series of standalone lectures brings in Data Science and Machine learning experts from the industry to talk about the "real-world" aspects and challenges of creating ML-based products. Each week a different guest lecturer introduces their company and its ML-based products and speaks in detail about various technical and product topics and challenges inherent to their work. Those talk aim to familiarize the community with the day-to-day challenges of data science work in different domains and introduce them to some of the leading tech companies in the field, their products and their unique features.

The talks are given to Y-DATA students at TAU campus as part of Fall semester curriculum. We open most of these lectures to our meetup community via zoom livestreaming.

Talk #4
26/11 12:15-13:15
Ofir Bibi: How you learn big things with little data

We are all used to saying "big data" and looking at given datasets, but in the real world, getting the right data to train on is not always easy and sometimes impossible. At Lightricks Research many of ML algorithms are deployed to the edge, which makes it even harder to use the data flowing through our models. In this talk Ofir will present several ways to overcome such challenges, including the use of self supervised learning, defining the important statistics for learning and the use of proxy data.

Bio:
Ofir is the VP of Research at Lightricks, a creative tools and market company with multiple apps and services and 40M+ MAUs. He is the research executive that built the department bringing magic to the Lightricks apps by getting state of the art ML research to the hands of consumers.