Search

We propose a domain-adapted reward model that works alongside an Offline A/B testing system for evaluating ranking models. This approach effectively measures reward for ranking model changes in large-scale Ads recommender systems, where model-free methods like IPS are not feasible. Our experiments demonstrate that the proposed technique outperforms both the vanilla IPS method and approaches using non-generalized reward models., Comment: Accepted at the CONSEQUENCES'24 workshop, co-located with ACM RecSys'24

It is often critical for prediction models to be robust to distributional shifts between training and testing data. From a causal perspective, the challenge is to distinguish the stable causal relationships from the unstable spurious correlations across shifts. We describe a causal transfer random forest (CTRF) that combines existing training data with a small amount of data from a randomized experiment to train a model which is robust to the feature shifts and therefore transfers to a new targeting distribution. Theoretically, we justify the robustness of the approach against feature shifts with the knowledge from causal learning. Empirically, we evaluate the CTRF using both synthetic data experiments and real-world experiments in the Bing Ads platform, including a click prediction task and in the context of an end-to-end counterfactual optimization system. The proposed CTRF produces robust predictions and outperforms most baseline methods compared in the presence of feature shifts., Comment: 9 pages, 7 figures, 2 tables, accepted to WSDM 2021

In classical causal inference, inferring cause-effect relations from data relies on the assumption that units are independent and identically distributed. This assumption is violated in settings where units are related through a network of dependencies. An example of such a setting is ad placement in sponsored search advertising, where the clickability of a particular ad is potentially influenced by where it is placed and where other ads are placed on the search result page. In such scenarios, confounding arises due to not only the individual ad-level covariates but also the placements and covariates of other ads in the system. In this paper, we leverage the language of causal inference in the presence of interference to model interactions among the ads. Quantification of such interactions allows us to better understand the click behavior of users, which in turn impacts the revenue of the host search engine and enhances user satisfaction. We illustrate the utility of our formalization through experiments carried out on the ad placement system of the Bing search engine.

In this paper, we propose an offline counterfactual policy estimation framework called Genie to optimize Sponsored Search Marketplace. Genie employs an open box simulation engine with click calibration model to compute the KPI impact of any modification to the system. From the experimental results on Bing traffic, we showed that Genie performs better than existing observational approaches that employs randomized experiments for traffic slices that have frequent policy updates. We also show that Genie can be used to tune completely new policies efficiently without creating risky randomized experiments due to cold start problem. As time of today, Genie hosts more than 10000 optimization jobs yearly which runs more than 30 Million processing node hours of big data jobs for Bing Ads. For the last 3 years, Genie has been proven to be the one of the major platforms to optimize Bing Ads Marketplace due to its reliability under frequent policy changes and its efficiency to minimize risks in real experiments., Comment: 20 pages, 13 figures

This paper introduces a new method for the session construction problem, which is the first main step of the web usage mining process. Through experiments, it is shown that when our new technique is used, it outperforms previous approaches in web usage mining applications such as next-page prediction.

Web usage mining is a type of web mining, which exploits data mining techniques to discover valuable information from navigation behavior of World Wide Web users. As in classical data mining, data preparation and pattern discovery are the main issues in web usage mining. The first phase of web usage mining is the data processing phase, which includes the session reconstruction operation from server logs. Session reconstruction success directly affects the quality of the frequent patterns discovered in the next phase. In reactive web usage mining techniques, the source data is web server logs and the topology of the web pages served by the web server domain. Other kinds of information collected during the interactive browsing of web site by user, such as cookies or web logs containing similar information, are not used. The next phase of web usage mining is discovering frequent user navigation patterns. In this phase, pattern discovery methods are applied on the reconstructed sessions obtained in the first phase in order to discover frequent user patterns. In this paper, we propose a frequent web usage pattern discovery method that can be applied after session reconstruction phase. In order to compare accuracy performance of session reconstruction phase and pattern discovery phase, we have used an agent simulator, which models behavior of web users and generates web user navigation as well as the log data kept by the web server., Comment: 19 pages, 6 figures