Differentially Private Matrix Factorization

Authors: Jingyu Hua, Chang Xia, Sheng Zhong

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We now evaluate the performance of our proposal based on two public datasets. The first one is reduced from the well-known Netflix dataset. ... The second one is Movie Lens 100k...So, our evaluations mainly use MF accuracy and communication overheads as the key metrics.
Researcher Affiliation Academia Jingyu Hua, Chang Xia, Sheng Zhong State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, China huajingyu@nju.edu.cn, changxia656569@gmail.com, zhongsheng@nju.edu.cn
Pseudocode No The paper describes computational steps and updating rules (e.g., equations 2, 3, 4, 5, 9, 12, 13) but does not present them in a structured pseudocode or algorithm block.
Open Source Code No The paper does not include any explicit statements about releasing source code or provide links to a code repository.
Open Datasets Yes The first one is reduced from the well-known Netflix dataset... The second one is Movie Lens 100k, which consists of 943 users ratings of 1682 movies.
Dataset Splits No The paper does not explicitly provide details about train/validation/test dataset splits with specific percentages, counts, or predefined split references. It evaluates 'prediction errors' on ratings, but the exact partitioning is not detailed for reproduction.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU, GPU models, memory, or specific cloud instances) used for running the experiments.
Software Dependencies No The paper describes algorithms (e.g., SGD-based MF) but does not list any specific software components or their version numbers (e.g., Python, PyTorch, TensorFlow, scikit-learn) required to reproduce the experiments.
Experiment Setup Yes The gain factor γ is 2 5. The parameters λ and µ in (1) are both set to 0.001. The dimension d of all the profile vectors is set to 50.