Personalized Adaptive Meta Learning for Cold-start User Preference Prediction
Authors: Runsheng Yu, Yu Gong, Xu He, Yu Zhu, Qingwen Liu, Wenwu Ou, Bo An10772-10780
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on Movie Lens, Book Crossing, and real-world production datasets reveal that our method outperforms the state-of-the-art methods dramatically for both the minor and major users. |
| Researcher Affiliation | Collaboration | Runsheng Yu,1 Yu Gong, 2 Xu He, 1 Yu Zhu , 2 Qingwen Liu,2 Wenwu Ou,2 Bo An 1 1 School of Computer Science and Engineering, Nanyang Technological University 2 Alibaba Group |
| Pseudocode | Yes | The pseudo-code can be found in Algs. 1. |
| Open Source Code | No | No explicit statement about providing open-source code for the described methodology was found, nor a direct link to a code repository for their work. The only link (https://pytorch.org/) refers to a third-party library. |
| Open Datasets | Yes | We use both open-source datasets as well as real-world production dataset to evaluate the performance of our methods in user imbalanced dataset, including Movie Lens-1M (Harper and Konstan 2015), Book Crossing (Ziegler et al. 2005) as well as production dataset (collected from Taobao e-commerce platforms, which is somewhat similar to (Zhao et al. 2018; Guo et al. 2019)). |
| Dataset Splits | Yes | We randomly split the users by 7 : 1 : 2 for training, validation, and testing. ... We separate the support and query sets for each user with a ratio of 80%:20% randomly. |
| Hardware Specification | Yes | All the experiments are done on a single Ge Force GTX 1080 Ti. |
| Software Dependencies | No | The paper mentions 'Pytorch' and 'ADAM' as the optimizer, and 'scipy.stats.ttest_ind' package for t-test, but does not provide specific version numbers for these software components. |
| Experiment Setup | No | The paper includes a section titled 'Experimental Setup' that describes datasets, user separation criteria, baselines, and general network architecture. However, it does not provide specific numerical values for critical hyperparameters such as batch size, number of epochs, initial learning rates, or other detailed optimizer settings. |