DeepLight: Deep Lightweight Feature Interactions for Accelerating CTR Predictions in Ad Serving

Click-through rate (CTR) prediction is a crucial task in online display advertising. The embedding-based neural networks have beenproposed to learn both explicit feature interactions through a shallow component and deep feature interactions using a deep neuralnetwork (DNN) component. These sophisticated models, however, slow down the prediction inference by at least hundreds of times. Toaddress the issue of significantly increased serving delay and high memory usage for ad serving in production, this paper presentsDeepLight: a framework to accelerate the CTR predictions in three aspects: 1) accelerate the model inference via explicitly searchinginformative feature interactions in the shallow component; 2) prune redundant layers and parameters at intra-layer and inter-layer levelin the DNN component; 3) promote the sparsity of the embedding layer to preserve the most discriminant signals. By combining theabove efforts, the proposed approach accelerates the model inference by 46X on Criteo dataset and 27X on Avazu dataset withoutany loss on the prediction accuracy. This paves the way for successfully deploying complicated embedding-based neural networksin production for ad serving.