Abstract:Collaborative filtering (CF) works have demonstrated the robust capabilities of Shallow Autoencoders on implicit feedback, showcasing highly competitive performance with other reasonable approaches (e.g., iALS and VAE-CF). However, despite their dual advantages of high performance and simple construction, EASE still exhibits several major shortcomings that must be addressed. To be more precise, the scalability of EASE is limited by the number of items, which determines the storage and inversion cost of a large dense matrix; the square-loss optimization objective does not consistently meet the recommendation task’s requirement for predicting personalized rankings, resulting in sub-optimal outcomes; the regularization coefficients are sensitive and require re-calibration with different datasets, leading to an exhaustive and time-consuming fine-tuning process. In order to address these obstacles, we propose a novel approach called Similarity-Structure Aware Shallow Autoencoder (AutoS 2 AE) that aims to enhance both recommendation accuracy and model efficiency. Our method introduces three similarity structures: Co-Occurrence, KNN, and NSW graphs, which replace the large dense matrix in EASE with a sparse structure, thus facilitating model compression. Additionally, we optimize the model by incorporating a low-rank training component into the matrix and applying a weighted square loss for improved ranking-oriented approximations. To automatically tune the hyperparameters, we further design two validation losses on the validation set for guidance and update the hyperparameters using the gradients of these validation losses. Both theoretical analyses regarding the introduction of similarity structures and empirical evaluations on multiple real-world datasets demonstrate the effectiveness of our proposed method, which significantly outperforms competing baselines.

Incorporating Collaborative Ranking Algorithm with Weighted Recursive Autoencoder for Item Recommendation.

Improving Recommendations With Collaborative Factors

Recommendation with Item Clustering Based Collaborative Filtering

A User-based and Item-based Collaborative Filtering Recommendation Algorithm

A Hybrid Recommendation Algorithm Based on Clustering and Collaborative Filtering

Collaborative Deep Ranking: A Hybrid Pair-Wise Recommendation Algorithm with Implicit Feedback

Hybrid Collaborative Recommendation via Dual-Autoencoder

Learning-Efficient Yet Generalizable Collaborative Filtering for Item Recommendation

Fusion Recommendation Algorithm Based on DeepWalk and Item-Based Collaborative Filtering

Representation Learning with Pair-wise Constraints for Collaborative Ranking

ILCR: Item-based Latent Factors for Sparse Collaborative Retrieval

Representation Learning with Collaborative Autoencoder for Personalized Recommendation

Attentive and Collaborative Deep Learning for Recommendation

Collaborative Autoencoder for Recommender Systems

Neural Reranking-Based Collaborative Filtering by Leveraging Listwise Relative Ranking Information

High-order autoencoder with data augmentation for collaborative filtering

Deep Collaborative Autoencoder for Recommender Systems: A Unified Framework for Explicit and Implicit Feedback.

Collaborative Knowledge Base Embedding for Recommender Systems

An Efficient Method for Autoencoder‐based Collaborative Filtering

Automated Sparse and Low-Rank Shallow Autoencoders for Recommendation

Hybrid collaborative recommendation of co-embedded item attributes and graph features