Embeddings paradigm shift: model training to vector similarity search
AI/ML Developer Advocate, Redis
Nava is Developer Advocate for Data Science and MLOps at Redis. She started her career in tech in an R&D Unit in the IDF, and later on was fortunate to be dealing with and championing transformational technologies in Cloud, Big Data and AI / ML, early at the beginning of each wave, in startups and in large global companies as Amdocs and Intel. Nava is also a mentor and judge at the MassChallenge accelerator, and the founder of LerGO, a cloud-based nonprofit EdTech venture.
Deep learning is considered the most successful field of AI research and learned vector embeddings is one of its most fascinating concepts. This talk will explore the shift we see today in using embeddings: From model training—as an input to downstream ML models, to semantic search—calculating the vector similarity between embeddings during production and discuss its profound implications.