On some of the largest commercial platforms, recommendations account for as much as 30 percent of revenue, which can translate into billions of dollars in sales. That’s why banks and insurance companies are using recommender systems to drive every action consumers take, from visiting a webpage to prioritizing which debt to pay off first. Recommenders also increase conversion by providing personalized messages to consumers, improving customer loyalty and satisfaction with the bank.
NVIDIA Merlin™ is an end-to-end recommender-on-GPU framework that provides fast feature engineering and high-training throughput to enable fast experimentation and production retraining of deep learning models. Merlin also enables low-latency, high-throughput, production inference to deliver personalized customer interactions with speed and accuracy.