Collaborative Recurrent Autoencoder: Recommend while Learning to Fill in the Blanks
Authors: Hao Wang, Xingjian SHI, Dit-Yan Yeung
NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on real-world datasets from different domains (Cite ULike and Netflix) show that, by jointly modeling the order-aware generation of sequences for the content information and performing CF for the ratings, CRAE is able to significantly outperform the state of the art on both the recommendation task based on ratings and the sequence generation task based on content information. |
| Researcher Affiliation | Academia | Hao Wang, Xingjian Shi, Dit-Yan Yeung Hong Kong University of Science and Technology {hwangaz,xshiab,dyyeung}@cse.ust.hk |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., specific repository link, explicit code release statement) for the source code. |
| Open Datasets | Yes | We use two datasets from different real-world domains. Cite ULike is from [20] with 5,551 users and 16,980 items (articles with text). Netflix consists of 407,261 users, 9,228 movies, and 15,348,808 ratings after removing users with less than 3 positive ratings (following [23], ratings larger than 3 are regarded as positive ratings). Please see Section 7 of the supplementary materials for details. |
| Dataset Splits | Yes | For the recommendation task, similar to [21, 23], P items associated with each user are randomly selected to form the training set and the rest is used as the test set. We evaluate the models when the ratings are in different degrees of density (P {1, 2, . . . , 5}). For each value of P, we repeat the evaluation five times with different training sets and report the average performance. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific software dependency details (e.g., library names with version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | In the experiments, we use 5-fold cross validation to find the optimal hyperparameters for CRAE and the baselines. For CRAE, we set α = 1, β = 0.01, K = 50, KW = 100. The wildcard denoising rate is set to 0.4. See Section 5.1 of the supplementary materials for details. |