A naukri.com initiative
Fb
1M
19
Image Credit: Fb
Sequence learning: A paradigm shift for personalized ads recommendations
- Meta’s ad recommendation engine, powered by deep learning recommendation models (DLRMs), has been instrumental in delivering personalized ads to people.
- Keys to success include thousands of human-engineered signals or features in the DLRM-based recommendation system.Foundational transformations along two dimensions that have addressed limitations of traditional DLRMs are event-based learning and learning from sequences.
- Meta's new system for ads recommendations uses sequence learning at its core. This necessitated a complete redesign of the ads recommendations system across data storage, feature input formats, and model architecture.
- Event-based features - EBFs - are the building blocks for the new sequence learning models. EBFs - an upgrade to traditional features - standardizes heterogeneous inputs to sequence learning models.
- An event model synthesizes event embeddings from event attributes. It learns embeddings for each attribute and uses linear compression to summarize them into a single event attributed-based embedding.
- Following the redesign to shift from sparse feature learning to event-based sequence learning, the next focus was scaling across two domains — scaling the sequence learning architecture and scaling event sequences to be longer and richer.
- Meta’s next-generation recommendation system’s ability to learn directly from event sequences to better understand people’s preferences is further enhanced with longer sequences and richer event attributes.
- The impact and future of sequence learning are widely adopted across Meta’s ads systems, resulting in gains in ad relevance and performance, efficient infrastructure, and accelerated research velocity.
- Going forward, the focus will be on further scaling event sequences by 100X, developing more efficient sequence modeling architectures like linear attention and state space models, key-value (KV) cache optimization, and multimodal enrichment of event sequences.
Read Full Article
Like
For uninterrupted reading, download the app