Guanyu Lin

Guanyu Lin

Auto-generated profile, based on publicly available information

Guanyu Lin

Carnegie Mellon University, Pittsburgh, PA, USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 August 2020 in IEEE Intelligent Systems
Verified
Journal Article
Closed access
GLGuanyu Lin
Guanyu Lin
FLFeng Liang
Feng Liang
WPWeike Pan
Weike Pan
ZMZhong Ming
Zhong Ming
Recommendation models have been widely embedded in various online services, while most of which are designed with the assumption that users' original behaviors are available in a central server. This may cause the privacy issue. As a response, we follow a recent work called federated collaborative filtering (FCF) for item recommendation with implicit feedback, and propose a novel and generic federated recommendation (FedRec) framework for rating prediction with explicit feedback. Specifically, we federate some basic and advanced factorization-based recommendation models both in batch style and in stochastic style. More importantly, in order to protect the private information of which items each user has rated, as well as not to significantly increase the computational and communication cost, we design two simple but effective strategies, i.e., user averaging and hybrid filling, in which some (instead of all) unrated items are randomly sampled and assigned with some virtual ratings accordingly. Empirical studies on two public datasets show the effectiveness of our FedRec in terms of the closeness of a federated model and an unfederated one, and the usefulness of the two filling strategies.

4 March 2024
Verified
Conference Paper
Open access
GLGuanyu Lin
Guanyu Lin
CGChen Gao
Chen Gao
YZYu Zheng
Yu Zheng
JCJianxin Chang
Jianxin Chang
YNYanan Niu
Yanan Niu
MWMeng Wang
Meng Wang
In modern recommender systems, sequential recommendation leverages chronological user behaviors to make effective next-item suggestions, which suffers from data sparsity issues, especially for new users. One promising line of work is the cross-domain recommendation, which trains models with data across multiple domains to improve the performance in data-scarce domains. Recent proposed cross-domain sequential recommendation models such as PiNet and DASL have a common drawback relying heavily on overlapped users in different domains, which limits their usage in practical recommender systems. In this paper, we propose a M ixed A ttention N etwork (MAN) with local and global attention modules to extract the domain-specific and cross-domain information. Firstly, we propose a local/global encoding layer to capture the domain-specific/cross-domain sequential pattern. Then we propose a mixed attention layer with item similarity attention, sequence-fusion attention, and group-prototype attention to capture the local/global item similarity, fuse the local/global item sequence, and extract the user groups across different domains, respectively. Finally, we propose a local/global prediction layer to further evolve and combine the domain-specific and cross-domain interests. Experimental results on two real-world datasets (each with two domains) demonstrate the superiority of our proposed model. Further study also illustrates that our proposed method and components are model-agnostic and effective, respectively. The code and data are available at https://github.com/Guanyu-Lin/MAN.

30 April 2023
Verified
Conference Paper
Open access
GLGuanyu Lin
Guanyu Lin
CGChen Gao
Chen Gao
YZYu Zheng
Yu Zheng
JCJianxin Chang
Jianxin Chang
YNYanan Niu
Yanan Niu
YLYong Li
Yong Li
Accurate user interest modeling is vital for recommendation scenarios. One of the effective solutions is the sequential recommendation that relies on click behaviors, but this is not elegant in the video feed recommendation where users are passive in receiving the streaming contents and return skip or no-skip behaviors. Here skip and no-skip behaviors can be treated as negative and positive feedback, respectively. With the mixture of positive and negative feedback, it is challenging to capture the transition pattern of behavioral sequence. To do so, FeedRec has exploited a shared vanilla Transformer, which may be inelegant because head interaction of multi-heads attention does not consider different types of feedback. In this paper, we propose Dual-interest Factorization-heads Attention for Sequential Recommendation (short for DFAR) consisting of feedback-aware encoding layer, dual-interest disentangling layer and prediction layer. In the feedback-aware encoding layer, we first suppose each head of multi-heads attention can capture specific feedback relations. Then we further propose factorization-heads attention which can mask specific head interaction and inject feedback information so as to factorize the relation between different types of feedback. Additionally, we propose a dual-interest disentangling layer to decouple positive and negative interests before performing disentanglement on their representations. Finally, we evolve the positive and negative interests by corresponding towers whose outputs are contrastive by BPR loss. Experiments on two real-world datasets show the superiority of our proposed method against state-of-the-art baselines. Further ablation study and visualization also sustain its effectiveness. We release the source code here: https://github.com/tsinghua-fib-lab/WWW2023-DFAR.

4 March 2024
Verified
Conference Paper
Open access
GLGuanyu Lin
Guanyu Lin
CGChen Gao
Chen Gao
YZYu Zheng
Yu Zheng
YLYinfeng Li
Yinfeng Li
JCJianxin Chang
Jianxin Chang
YLYong Li
Yong Li
Modern personalized recommendation services often rely on user feedback, either explicit or implicit, to improve the quality of services. Explicit feedback refers to behaviors like ratings, while implicit feedback refers to behaviors like user clicks. However, in the scenario of full-screen video viewing experiences like Tiktok and Reels, the click action is absent, resulting in unclear feedback from users, hence introducing noises in modeling training. Existing approaches on de-noising recommendation mainly focus on positive instances while ignoring the noise in a large amount of sampled negative feedback. In this paper, we propose a meta-learning method to annotate the unlabeled data from loss and gradient perspectives, which considers the noises in both positive and negative instances. Specifically, we first propose anInverse Dual Loss (IDL) to boost the true label learning and prevent the false label learning. Then we further propose anInverse Gradient (IG) method to explore the correct updating gradient and adjust the updating based on meta-learning. Finally, we conduct extensive experiments on both benchmark and industrial datasets where our proposed method can significantly improve AUC by 9.25% against state-of-the-art methods. Further analysis verifies the proposed inverse learning framework is model-agnostic and can improve a variety of recommendation backbones. The source code, along with the best hyper-parameter settings, is available at this link: https://github.com/Guanyu-Lin/InverseLearning.