Be Aware of the Neighborhood Effect: Modeling Selection Bias under Interference

Authors: Haoxuan Li, Chunyuan Zheng, Sihao Ding, Peng Wu, Zhi Geng, Fuli Feng, Xiangnan He

ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive semi-synthetic and real-world experiments are conducted to demonstrate the effectiveness of the proposed methods.
Researcher Affiliation Academia 1Peking University 2University of Science and Technology of China 3Beijing Technology and Business University hxli@stu.pku.edu.cn dsihao@mail.ustc.edu.cn {zhengchunyuan99, fulifeng93, xiangnanhe}@gmail.com {pengwu, zhigeng}@btbu.edu.cn
Pseudocode Yes Algorithm 1: The Proposed Propensity Learning Algorithm, Algorithm 2: The Proposed N-IPS Learning Algorithm, Algorithm 3: The Proposed N-DR-JL Learning Algorithm, Algorithm 4: The Proposed N-MRDR-JL Learning Algorithm
Open Source Code Yes Our codes and datasets are available at https://github.com/haoxuanli-pku/ICLR24-Interference.
Open Datasets Yes We conduct semi-synthetic experiments using the Movie Lens 100K1 (ML-100K) dataset..., Coat 3 contains 6,960 MNAR ratings and 4,640 missing-at-random (MAR) ratings., Yahoo! R34 contains 311,704 MNAR ratings and 54,000 MAR ratings., Kuai Rec5 (Gao et al., 2022) is a public large-scale industrial dataset, which contains 4,676,570 video watching ratio records from 1,411 users for 3,327 videos.
Dataset Splits No The paper mentions 'training the prediction model' and 'test data' but does not explicitly provide specific training/validation/test dataset splits (e.g., percentages or counts) or reference standard predefined splits for its experimental setup.
Hardware Specification Yes For all experiments, we use NVIDIA Ge Force RTX 3090 as the computing resource.
Software Dependencies No All the experiments are implemented on Py Torch with Adam as the optimizer.
Experiment Setup Yes We tune the learning rate in {0.005, 0.01, 0.05, 0.1} and weight decay in [1e 6, 1e 2]. We tune bandwidth value in {40, 45, 50, 55, 60} for Coat, {1000, 1500, 2000, 2500, 3000} for Yahoo! R3 and {100, 150, 200, 250, 300} for Kuai Rec.