Projective Quadratic Regression for Online Learning

Authors: Wenye Ma5093-5100

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate the performance of the proposed PQR model in terms of accuracy and efficiency by comparing with the state-of-the-art methods.
Researcher Affiliation Industry Wenye Ma Tencent wenyema@tencent.com
Pseudocode Yes Algorithm 1: Follow-The-Regularized-Leader
Open Source Code No The paper does not provide an explicit statement about releasing the source code for the described methodology or a direct link to a code repository.
Open Datasets Yes For the online rating prediction task, we use the typical Movie Lens datasets, including Movie Lens-100K (ML100K), Movie Lens-1M (ML1M) and Movie Lens-10M (ML10M); For the online binary classification task, we select high-dimensional sparse datasets including Avazu, Criteo3, and KDD20124. These three high-dimensional datasets are preprocessed and can be downloaded from the LIBSVM website5.
Dataset Splits Yes All these datasets are randomly separated into train(80%), validation(10%) and test(10%) sets.
Hardware Specification No The paper does not specify the hardware (e.g., CPU, GPU models, or cloud computing instances) used for running the experiments.
Software Dependencies No The paper mentions using 'FTRL-Proximal algorithm' and 'online gradient descent' but does not specify version numbers for any software, libraries, or frameworks used in the implementation.
Experiment Setup Yes The rank of FM is selected from the set of {2, 4, 8, 16, 32, 64, 128, 256, 512}. The coefficient of regularization term and the learning rate are tuned in a range [0.0001, 10].