Arbitrary Distribution Modeling with Censorship in Real-Time Bidding Advertising

TL;DR

This paper introduces a novel framework called Arbitrary Distribution Modeling (ADM) with a new loss function, Neighborhood Likelihood Loss (NLL), to accurately predict winning price distributions in real-time bidding without assuming specific distribution forms.

Contribution

The paper proposes a flexible, efficient method for modeling winning price distributions in RTB that handles censorship and does not rely on predefined distribution assumptions.

Findings

ADM outperformed baselines on algorithm and business metrics

The method demonstrated effectiveness on real-world and production datasets

It achieved good yield in the production environment

Abstract

The purpose of Inventory Pricing is to bid the right prices to online ad opportunities, which is crucial for a Demand-Side Platform (DSP) to win advertising auctions in Real-Time Bidding (RTB). In the planning stage, advertisers need the forecast of probabilistic models to make bidding decisions. However, most of the previous works made strong assumptions on the distribution form of the winning price, which reduced their accuracy and weakened their ability to make generalizations. Though some works recently tried to fit the distribution directly, their complex structure lacked efficiency on online inference. In this paper, we devise a novel loss function, Neighborhood Likelihood Loss (NLL), collaborating with a proposed framework, Arbitrary Distribution Modeling (ADM), to predict the winning price distribution under censorship with no pre-assumption required. We conducted experiments on two real-world experimental datasets and one large-scale, non-simulated production dataset in our system. Experiments showed that ADM outperformed the baselines both on algorithm and business metrics. By replaying historical data of the production environment, this method was shown to lead to good yield in our system. Without any pre-assumed specific distribution form, ADM showed significant advantages in effectiveness and efficiency, demonstrating its great capability in modeling sophisticated price landscapes.