Abstract
Training models to predict click and order targets at the same time. For better user satisfaction and business effectiveness, multitask learning is one of the most important methods in e-commerce. Some existing researches model user representation based on historical behaviour sequence to capture user interests. It is often the case that user interests may change from their past routines. However, multi-perspective attention has broad horizon, which covers different characteristics of human reasoning, emotions, perception, attention, and memory. In this paper, we attempt to introduce the multi-perspective attention and sequence behaviour into multitask learning. Our proposed method offers better understanding of user interest and decision. To achieve more flexible parameter sharing and maintaining the special feature advantage of each task, we improve the attention mechanism at the view of expert interactive. To the best of our knowledge, we firstly propose the implicit interaction mode, the explicit hard interaction mode, the explicit soft interaction mode, and the data fusion mode in multitask learning. We do experiments on public data and lab medical data. The results show that our model consistently achieves remarkable improvements to the state-of-the-art method.
Highlights
In the real world, there are some scenarios for multitask learning
In the e-commerce field, we need to increase click through rate (CTR) and order conversion rate (CVR) at the same time
Compared with the baseline multi-gate mixture of experts (MMOE), ESMM, and CGC, we demonstrate the effectiveness of our approach on Ali-CCP public dataset
Summary
There are some scenarios for multitask learning. In the e-commerce field, we need to increase click through rate (CTR) and order conversion rate (CVR) at the same time. E model applies user behaviour sequence feature in multitask learning scenarios. According to the multitask learning framework, Shao et al [17] introduce attention map convolutional layer to mine the bilateral high-order feature graph from user and commodity. Based on the knowledge graph, Yu et al [19] propose a multitask feature learning method using the knowledge graph to calculate the embedding vector assist the recommendation task . The above studies in multitask learning are based on feature engineering and knowledge representation, without introducing multi-perspective attention. (3) To the best of our knowledge, we are the first to introduce the expert-level multi-head self-attention into multitask learning and get better effectiveness. (4) We design the time-space sequence feature into multitask learning and improve the loss function, which can support multiple-source datasets. (5) We conduct extensive experiments on Ali-CCP data and confirm the superiority of our proposed model over representative state-of-the-art method
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
