A Nonparametric Delayed Feedback Model for Conversion Rate Prediction

Abstract

Predicting conversion rates (CVRs) in display advertising (e.g., predicting the proportion of users who purchase an item (i.e., a conversion) after its corresponding ad is clicked) is important when measuring the effects of ads shown to users and to understanding the interests of the users. There is generally a time delay (i.e., so-called delayed feedback) between the ad click and conversion. Owing to the delayed feedback, samples that are converted after an observation period may be treated as negative. To overcome this drawback, CVR prediction assuming that the time delay follows an exponential distribution has been proposed. In practice, however, there is no guarantee that the delay is generated from the exponential distribution, and the best distribution with which to represent the delay depends on the data. In this paper, we propose a nonparametric delayed feedback model for CVR prediction that represents the distribution of the time delay without assuming a parametric distribution, such as an exponential or Weibull distribution. Because the distribution of the time delay is modeled depending on the content of an ad and the features of a user, various shapes of the distribution can be represented potentially. In experiments, we show that the proposed model can capture the distribution for the time delay on a synthetic dataset, even when the distribution is complicated. Moreover, on a real dataset, we show that the proposed model outperforms the existing method that assumes an exponential distribution for the time delay in terms of conversion rate prediction.

Addressing delayed feedback for continuous training with neural networks in CTR prediction

“…Even though most approaches disregard the time delay information that is available (i.e. time that has elapsed since the impression and time until the user engages with the ad), some of them leverage time-delay by jointly training a delay model along with a CPC or CPA model [3,28]. In the following sections, we describe five different methods, which constitute potential solutions to the FN problem.…”

“…4 illustrates that the time-to-click also follows an exponential distribution, which renders this model an appropriate solution. As an extension of the model presented in [3], [28] suggested a non-parametric delayed feedback model (NoDeF) to capture the time delay without assuming any parametric distributions, like exponential or Weibull. This model assumes a hidden variable for each sample, which indicates whether this action will eventually lead to conversion.…”

Follow the Prophet: Accurate Online Conversion Rate Prediction in the Face of Delayed Feedback

“…Chapelle [5] proposed to apply survival time analysis with a delayed feedback model (DFM) to estimate the delay under an assumption of exponential distribution. Yoshikawa and Imai [26] extended DFM to a non-parametric model (NoDeF). Yasui et al [25] regarded the delayed feedback as a data shift and proposed to use an importance weighting approach (FSIW) to handle different distribution between test data and observed data.…”

Stochastic bandits with arm-dependent delays

“…The problem of learning with delayed feedback is ubiquitous in a wide range of applications, including universal portfolios in finance (Cover, 2011), online advertising (Chapelle, 2014) and e-commerce (Yoshikawa and Imai, 2018). Therefore, there is a large body of theoretical results designed under different scenarios and assumptions on the delays.…”