You are only as good as your embeddings - how to train high quality models for production vector search

This talk covers practicalities of training embedding models for production (multi-modal) vector search. Topics will span data, training, and evaluation. More specifically the talk will cover:

• How to optimize model training for business objectives by leveraging historic search-result interactions.
• The importance of data quality, including how to think about query-document mixes, duplicates, and query coverage.
• Leveraging existing search results for behavior retention and model regularization.
• Adapting strategies from recommendation systems, such as bias terms, linear re-ranking and query-result interaction matrices.
• Loss functions, base models for fine-tuning, and key hyperparameter considerations.
• Production-aware training techniques, including optimization for vector databases, vector fusing, and binary/truncation-aware training.
• Efficient updating without re-indexing.
• Transitioning from offline to online A/B testing, with a focus on novelty-based splits.

Robertson Taylor

Marqo

Robertson is a solutions engineer with experience working on high-volume data solutions. At Marqo, Robertson helps businesses create fine-tuned embedding models and scalable retrieval solutions. His day to day is using embeddings to handle problems like product retrieval, content classification, and real-time recommendations, and help customers understand ways they can use existing models that they might not be thinking about.