Lift-Based Bidding in Ad Selection

Authors: Jian Xu, Xuhui Shao, Jianjie Ma, Kuang-chih Lee, Hang Qi, Quan Lu

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We set up A/B test experiments on Yahoo’s Demand-Side Platform. ... The results shown in Table 2, 3, 4 and 5 backed up our claims and methods.
Researcher Affiliation Industry Jian Xu , Xuhui Shao, Jianjie Ma, Kuang-chih Lee, Hang Qi, Quan Lu Touch Pal Inc., 1172 Castro St, Mountain View, CA 94040 Yahoo Inc., 701 First Ave, Sunnyvale, CA 94089 jian.xu@cootek.cn, {xshao,jianma,kclee,hangqi,qlu}@yahoo-inc.com
Pseudocode No No structured pseudocode or algorithm blocks were found in the paper.
Open Source Code No The paper does not contain any statement about releasing source code or a link to a code repository for the methodology described.
Open Datasets No We set up A/B test experiments on Yahoo’s Demand-Side Platform. We selected five advertisers to participate in the test.
Dataset Splits No Our task is to train a generic AR prediction model ˆP to give AR estimations for both cases when an ad is shown or not shown. ... We set up A/B test experiments on Yahoo’s Demand-Side Platform. We first randomly split users into three equal-sized groups.
Hardware Specification No No specific hardware details (like GPU/CPU models, memory, or cloud instances) used for running experiments were mentioned in the paper.
Software Dependencies No Once the training samples are gathered, we train a Gradient-Boosting-Decision-Tree (GBDT) model to predict the rank order and then calibrate using isotonic regression to translate a GBDT score to an AR. Please note that we utilize our in-house GBDT tool with distributed training capability for modeling; however, other proper machine learning models can also be applied.
Experiment Setup No Our task is to train a generic AR prediction model... we train a Gradient-Boosting-Decision-Tree (GBDT) model to predict the rank order and then calibrate using isotonic regression to translate a GBDT score to an AR.