TDR-CL: Targeted Doubly Robust Collaborative Learning for Debiased Recommendations
Authors: Haoxuan Li, Yan Lyu, Chunyuan Zheng, Peng Wu
ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, following the previous studies (Schnabel et al., 2016; Wang et al., 2019; Saito, 2020; Guo et al., 2021), we aim to answer the following research question (RQ) on the semi-synthetic datasets: RQ1. Does the proposed TDR estimator in estimating the ideal loss have both the statistical properties of lower bias and variance in the presence of selection bias? (...) In this section, we conduct experiments to evaluate the proposed methods on two real-world benchmark datasets containing missing-at-random (MAR) ratings. |
| Researcher Affiliation | Academia | 1Peking University 2University of California, San Diego 3Beijing Technology and Business University |
| Pseudocode | Yes | Algorithm 1: The Proposed Targeted Doubly Robust Collaborative Learning, TDR-CL |
| Open Source Code | Yes | Code is provided in Supplementary Materials to reproduce the experimental results. |
| Open Datasets | Yes | Movie Lens 100K1 (ML-100K) is a dataset of 100,000 missing-not-at-random (MNAR) ratings from 943 users and 1,682 movies collected from movie recommendation ratings. Movie Lens 1M2 (ML-1M) is a larger dataset of 1,000,209 MNAR ratings from 6,040 users and 3,952 movies. (...) Coat Shopping3 has 4,640 MAR and 6,960 MNAR ratings of 290 users to 300 Coats. Music! R34 has 54,000 MAR and 311,704 MNAR ratings of 15,400 users to 1,000 songs. (URLs provided in footnotes) |
| Dataset Splits | No | The paper states, 'After finding out the best configuration on the validation set, we evaluate the trained models on the MAR test set.' and refers to 'MAR test set' elsewhere. However, it does not explicitly provide the specific percentages or counts for training, validation, and test dataset splits. |
| Hardware Specification | Yes | Experiments are conducted using NVIDIA Ge Force RTX 3090. |
| Software Dependencies | No | The paper mentions 'Adam is utilized as the optimizer' and discusses tuning parameters like learning rate and batch size. However, it does not provide specific version numbers for programming languages (e.g., Python) or libraries (e.g., PyTorch, TensorFlow, scikit-learn) used in the implementation. |
| Experiment Setup | Yes | We tune the learning rate in {0.001, 0.005, 0.01, 0.05, 0.1}, weight decay in [1e 6, 1e 2] at 10x multiplicative ratio, and batch size in {128, 256, 512, 1024, 2048} for Coat and {1024, 2048, 4096, 8192, 16384} for Music! R3. Specifically for the propensity training, we tune the clipping threshold in {0.05, 0.10, 0.15, 0.20}. |