AdaSparse: Learning Adaptively Sparse Structures for Multi-Domain Click-Through Rate Prediction
This work addresses computational efficiency and generalization issues in multi-domain CTR prediction for recommendation and advertising systems, representing an incremental improvement over existing methods.
The paper tackles the challenge of improving generalization across domains in multi-domain click-through rate prediction under limited training data and high computational complexity, proposing AdaSparse which learns adaptively sparse structures for each domain and shows significant performance improvements in offline and online experiments.
Click-through rate (CTR) prediction is a fundamental technique in recommendation and advertising systems. Recent studies have proved that learning a unified model to serve multiple domains is effective to improve the overall performance. However, it is still challenging to improve generalization across domains under limited training data, and hard to deploy current solutions due to their computational complexity. In this paper, we propose a simple yet effective framework AdaSparse for multi-domain CTR prediction, which learns adaptively sparse structure for each domain, achieving better generalization across domains with lower computational cost. In AdaSparse, we introduce domain-aware neuron-level weighting factors to measure the importance of neurons, with that for each domain our model can prune redundant neurons to improve generalization. We further add flexible sparsity regularizations to control the sparsity ratio of learned structures. Offline and online experiments show that AdaSparse outperforms previous multi-domain CTR models significantly.