Local Collaborative Autoencoders

Choi, Minjin; Jeong, Yoonki; Lee, Joon Seok; Lee, Jongwuk

doi:10.1145/3437963.3441808

Cited by 16 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Random pruning is a simple baseline, i.e., the parameters are randomly removed under the given compression ratio. This result indicates that the learned relationship between items is disentangled with the locality, as observed in existing studies [4,[27][28][29][30]. Owing to this valuable property, the compressed S-Walk is memory-efficient, suitable for low-resource devices, e.g., mobile and embedded applications.…”

Section: Evaluation Of Scalabilitysupporting

confidence: 71%

S-Walk

Choi

Kim

Lee

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

Self Cite

View full text Add to dashboard Cite

Session-based recommendation (SR) predicts the next items from a sequence of previous items consumed by an anonymous user. Most existing SR models focus only on modeling intra-session characteristics but pay less attention to inter-session relationships of items, which has the potential to improve accuracy. Another critical aspect of recommender systems is computational efficiency and scalability, considering practical feasibility in commercial applications. To account for both accuracy and scalability, we propose a novel session-based recommendation with a random walk, namely S-Walk. Precisely, S-Walk effectively captures intra-and inter-session correlations by handling high-order relationships among items using random walks with restart (RWR). By adopting linear models with closed-form solutions for transition and teleportation matrices that constitute RWR, S-Walk is highly efficient and scalable. Extensive experiments demonstrate that S-Walk achieves comparable or state-of-the-art performance in various metrics on four benchmark datasets. Moreover, the model learned by S-Walk can be highly compressed without sacrificing accuracy, conducting two or more orders of magnitude faster inference than existing DNN-based models, making it suitable for large-scale commercial systems.

show abstract

Section: Evaluation Of Scalabilitysupporting

confidence: 71%

S-Walk

Choi

Kim

Lee

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…Randomly divide the training data will destroy the collaborative information which could damage the recommendation performance. One promising approach is using community detection or clustering methods [15,23,34]. However, due to the underlying structure of real-world data, these methods may lead to highly unbalanced data partition.…”

Section: Challenges Of Recommendation Unlearningmentioning

confidence: 99%

“…As we have mentioned before, the data used for recommendation tasks usually contains rich collaborative information. To preserve the collaborative information, one promising approach is to rely on community detection and clustering methods [15,23,34]. However, direct application of them may lead to highly unbalanced data partition.…”

Section: Balanced Data Partitionmentioning

confidence: 99%

“…To keep the collaborative information of the data, we design three data partition strategies based on the similarities of users, items, and interactions, respectively. Different from traditional community detection and clustering methods [15,23,34], our data partition strategies are designed to achieve balanced partition so that the unlearning efficiency will not be affected by the unbalanced shard size. In addition, considering that recommender systems usually face a variety of users and items, submodels should have different contributions for the prediction of different user-item pairs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Recommendation Unlearning

Chen¹,

Sun²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recommender systems provide essential web services by learning users' personal preferences from collected data. However, in many cases, systems also need to forget some training data. From the perspective of privacy, users desire a tool to erase the impacts of their sensitive data from the trained models. From the perspective of utility, if a system's utility is damaged by some bad data, the system needs to forget such data to regain utility. While unlearning is very important, it has not been well-considered in existing recommender systems. Although there are some researches have studied the problem of machine unlearning, existing methods can not be directly applied to recommendation as they are unable to consider the collaborative information.In this paper, we propose RecEraser, a general and efficient machine unlearning framework tailored to recommendation tasks. The main idea of RecEraser is to divide the training set into multiple shards and train submodels with these shards. Specifically, to keep the collaborative information of the data, we first design three novel data partition algorithms to divide training data into balanced groups. We then further propose an adaptive aggregation method to improve the global model utility. Experimental results on three public benchmarks show that RecEraser can not only achieve efficient unlearning but also outperform the state-of-the-art unlearning methods in terms of model utility. The source code can be found at https://github.com/chenchongthu/Recommendation-Unlearning CCS CONCEPTS• Information systems → Recommender systems; • Security and privacy → Privacy protections.

show abstract

“…Collaborative filtering (CF) [1,7,22,36] is the most prevalent technique for building commercial recommender systems. CF typically utilizes two types of user feedback: explicit and implicit feedback.…”

Section: Introductionmentioning

confidence: 99%

Bilateral Self-unbiased Learning from Biased Implicit Feedback

Lee¹,

Park²,

Lee³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Implicit feedback has been widely used to build commercial recommender systems. Because observed feedback represents users' click logs, there is a semantic gap between true relevance and observed feedback. More importantly, observed feedback is usually biased towards popular items, thereby overestimating the actual relevance of popular items. Although existing studies have developed unbiased learning methods using inverse propensity weighting (IPW) or causal reasoning, they solely focus on eliminating the popularity bias of items. In this paper, we propose a novel unbiased recommender learning model, namely BIlateral SElf-unbiased Recommender (BISER), to eliminate the exposure bias of items caused by recommender models. Specifically, BISER consists of two key components: (i) self-inverse propensity weighting (SIPW) to gradually mitigate the bias of items without incurring high computational costs; and (ii) bilateral unbiased learning (BU) to bridge the gap between two complementary models in model predictions, i.e., user-and item-based autoencoders, alleviating the high variance of SIPW. Extensive experiments show that BISER consistently outperforms state-of-the-art unbiased recommender models over several datasets, including Coat, Yahoo! R3, MovieLens, and CiteULike.

show abstract

Local Collaborative Autoencoders

Cited by 16 publications

References 33 publications

S-Walk

S-Walk

Recommendation Unlearning

Bilateral Self-unbiased Learning from Biased Implicit Feedback

Contact Info

Product

Resources

About