Self-supervised learning (SSL) recently has achieved outstanding success on recommendation. By setting up an auxiliary task (either predictive or contrastive), SSL can discover supervisory signals from the raw data without human annotation, which greatly mitigates the problem of sparse user-item interactions. However, most SSL-based recommendation models rely on general-purpose auxiliary tasks, e.g., maximizing correspondence between node representations learned from the original and perturbed interaction graphs, which are explicitly irrelevant to the recommendation task. Accordingly, the rich semantics reflected by social relationships and item categories, which lie in the recommendation data-based heterogeneous graphs, are not fully exploited. To explore recommendation-specific auxiliary tasks, we first quantitatively analyze the heterogeneous interaction data and find a strong positive correlation between the interactions and the number of user-item paths induced by meta-paths. Based on the finding, we design two auxiliary tasks that are tightly coupled with the target task (one is predictive and the other one is contrastive) towards connecting recommendation with the self-supervision signals hiding in the positive correlation. Finally, a model-agnostic DUal-Auxiliary Learning (DUAL) framework which unifies the SSL and recommendation tasks is developed. The extensive experiments conducted on three real-world datasets demonstrate that DUAL can significantly improve recommendation, reaching the state-of-the-art performance.
翻译:最近,自监督学习(SSL)在建议上取得了杰出的成功。通过设置辅助任务(预测性或对比性),SSL可以在没有人类批注的情况下从原始数据中发现监督信号,这大大缓解了用户-项目互动稀少的问题。然而,大多数基于SSL的建议模型都依赖一般用途辅助任务,例如,尽量扩大从原始和扰动互动图中获取的节点代表之间的对应关系,这显然与建议任务无关。因此,社会关系和项目类别中反映的丰富的语义没有被充分利用,这些社会关系和项目类别存在于基于建议的数据混杂图中。为了探索针对具体建议的辅助任务,我们首先从数量上分析各种互动数据,发现互动与由元路径引发的用户-项目路径数量之间强有力的正相关关系。根据调查结果,我们设计了两个与目标任务(一个是预测性的,另一个是对比性的)紧密结合的辅助任务,将建议与隐藏在正相关文件中的自我监督信号连接起来。最后,我们从模型-不可辨别的图-DUI-SL-SAL-SAL 大量展示了SAL-SAL-SAL-SAL-SAL-SAL-SAL-SAL-SAL-I-SAL-SAL-T-T-T-SAL-SAL-T-T-T-T-T-T-T-T-ATIV-T-T-T-T-T-T-T-ATIV-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-Q-ATIV-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-Q-Q-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T