PPDL：一种融合物理先验与深度学习的工业级用户留存率预测框架
PPDL: A Real-world Industrial User Retention Ratio Forecasting Framework Integrating Physical Priors with Deep Learning
浏览论文内容
中文总结 AI 辅助
针对多渠道用户获取中留存率预测的渠道异质性、衰减饱和趋势和短回看窗口挑战，提出融合Weibull物理先验与深度学习的PPDL框架，通过趋势残差分解、渠道嵌入和多尺度趋势惩罚损失，在工业数据上显著优于现有方案。
中文摘要 AI 辅助
在多渠道付费用户获取中，早期准确预测渠道层面的用户留存对于优化预算分配至关重要。用户留存曲线呈现出显著的时间模式：最初的高流失期过渡到长期稳定期。这种模式进一步以季节性导致的规律性波动为特征，并表现出高度的序列自相关性。这些内在属性使得此类曲线非常适合在时间序列预测框架内进行分析。然而，预测大规模短视频平台的用户留存率面临三大挑战：渠道间的显著异质性、先衰减后饱和的明显全局趋势，以及较短的回看窗口。为应对这些挑战，我们提出了PPDL，一种将物理先验与深度学习相结合的新型预测框架。我们首先引入趋势-残差分解组件。趋势使用Weibull分布建模，其参数通过多层感知机（MLP）学习。其次，对于残差分量，我们在深度学习主干之上设计了一个辅助嵌入模块，以保持渠道身份感知。最后，为了增强模型对趋势的敏感性，我们设计了一种多尺度趋势惩罚损失函数。所提出的PPDL方法通过在工业规模数据集上的全面实验得到验证，涵盖了三个应用，每个应用平均有30多个渠道。实验结果表明，PPDL在不同主干网络上均取得了改进，并显著优于现有的在线解决方案。
英文摘要
In multi-channel paid user acquisition, early and accurate prediction of user retention at the channel level is crucial for optimizing budget allocation. User retention curves display a pronounced temporal pattern: an initial period of high churn transitions into long-term stability. This pattern is further characterized by regular fluctuations attributable to seasonality and exhibits high serial autocorrelation. These intrinsic properties make such curves highly suitable for analysis within a time-series forecasting framework. However, forecasting user retention ratio for large-scale short-video platform faces three major challenges: significant heterogeneity across channels, pronounced global trend of decay followed by saturation, and short look-back windows. To address these challenges, we propose PPDL, a novel forecasting framework that integrates physical priors with deep learning. We first introduce a trend-residual decomposition component. The trend is modeled using the Weibull distribution, whose parameters are learned via a Multilayer Perceptron (MLP). Secondly, for the residual component, we design an auxiliary embedding module on top of a deep learning backbone to maintain the channel identity awareness. Finally, to enhance the model's sensitivity to trends, we design a Multiscale Trend-penalized loss function. The proposed approach PPDL is validated through comprehensive experiments on industrial-scale datasets, covering three applications with an average of 30+ channels each. Experimental results show that PPDL achieves improvements across different backbones and significantly outperforms existing online solutions.
发表机构
- Zhejiang Normal University（浙江师范大学）
- Douyin Group（抖音集团）
- Shanghai Jiao Tong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。
