Learning Explicit User Interest Boundary for Recommendation

@article{Zhuo2022LearningEU,
  title={Learning Explicit User Interest Boundary for Recommendation},
  author={Jianhuan Zhuo and Qiannan Zhu and Yinliang Yue and Yuhong Zhao},
  journal={Proceedings of the ACM Web Conference 2022},
  year={2022}
}
The core objective of modelling recommender systems from implicit feedback is to maximize the positive sample score sp and minimize the negative sample score sn, which can usually be summarized into two paradigms: the pointwise and the pairwise. The pointwise approaches fit each sample with its label individually, which is flexible in weighting and sampling on instance-level but ignores the inherent ranking property. By qualitatively minimizing the relative score sn − sp, the pairwise… 

References

SHOWING 1-10 OF 41 REFERENCES
Representation Learning and Pairwise Ranking for Implicit Feedback in Recommendation Systems
TLDR
A novel ranking framework for collaborative filtering with the overall aim of learning user preferences over items by minimizing a pairwise ranking loss is proposed and it is demonstrated that the approach is very competitive with the best state-of-the-art collaborative filtering techniques proposed for implicit feedback.
Deep generative ranking for personalized recommendation
TLDR
A deep generative ranking (DGR) model under the Wasserstein autoencoder framework is proposed and experimental results demonstrate that DGR consistently benefit the recommendation system in ranking estimation task, especially for the near-cold-start-users.
Improving pairwise learning for item recommendation from implicit feedback
TLDR
The experiments indicate that the proposed adaptive sampler improves the state-of-the art learning algorithm largely in convergence without negative effects on prediction quality or iteration runtime.
Symmetric Metric Learning with Adaptive Margin for Recommendation
TLDR
A novel Symmetic Metric Learning with adaptive margin (SML) is proposed, which symmetically introduces a positive item-centric metric which maintains closer distance from positive items to user, and push the negative items away from the positive items at the same time.
Top-N Recommendation with Counterfactual User Preference Simulation
TLDR
This paper proposes to reformulate the recommendation task within the causal inference framework, which enables us to counterfactually simulate user ranking-based preferences to handle the data scarce problem.
SimpleX: A Simple and Strong Baseline for Collaborative Filtering
TLDR
It is shown that the choice of loss function as well as negative sampling ratio is equivalently important, and a simple unified CF model, dubbed SimpleX is proposed, which can surpass most sophisticated state-of-the-art models by a large margin.
BPR: Bayesian Personalized Ranking from Implicit Feedback
TLDR
This paper presents a generic optimization criterion BPR-Opt for personalized ranking that is the maximum posterior estimator derived from a Bayesian analysis of the problem and provides a generic learning algorithm for optimizing models with respect to B PR-Opt.
Reinforced Negative Sampling for Recommendation with Exposure Data
TLDR
This work designs a novel RNS method (short for Reinforced Negative Sampler) that generates exposure-alike negative instances through feature matching technique instead of directly choosing from exposure data, and is able to integrate user preference signals in exposure data and hard negatives.
Efficient Neural Matrix Factorization without Sampling for Recommendation
TLDR
This work derives three new optimization methods through rigorous mathematical reasoning, which can efficiently learn model parameters from the whole data with a rather low time complexity, and presents a general framework named ENMF, short for Efficient Neural Matrix Factorization.
Collaborative Translational Metric Learning
TLDR
This paper proposes TransCF, a method that outperforms several state-of-the-art methods for top-N recommendation on seven real-world data by up to 17% in terms of hit ratio and conducts extensive qualitative evaluations on the translation vectors learned by the method.
...
1
2
3
4
5
...