An Input-aware Factorization Machine for Sparse Prediction

Yu, Yanfang; Wang, Zhen; Yuan, Bo

doi:10.24963/ijcai.2019/203

Cited by 34 publications

(22 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, input-dependent models have shown effectiveness in various domains, such as language modeling [27], [28] and computer vision [29], [30]. In recommendation, IFM [31] and DIFM [32] are presented to re-weight the representations of features and weights for different input instances before performing feature interactions. Inspired by these studies, we design a dynamic transformer encoder which performs an individual attention network on the self-attention layer and enables modeling user-specific intra-item patterns unveiled by the user intentions.…”

Section: Related Workmentioning

confidence: 99%

Attention over Self-attention:Intention-aware Re-ranking with Dynamic Transformer Encoders for Recommendation

Lin¹,

Zang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Re-ranking models refine the item recommendation list generated by the prior global ranking model with intra-item relationships. However, most existing re-ranking solutions refine recommendation list based on the implicit feedback with a shared re-ranking model, which regrettably ignore the intra-item relationships under diverse user intentions. In this paper, we propose a novel Intention-aware Re-ranking Model with Dynamic Transformer Encoder (RAISE), aiming to perform user-specific prediction for each target user based on her intentions. Specifically, we first propose to mine latent user intentions from text reviews with an intention discovering module (IDM). By differentiating the importance of review information with a co-attention network, the latent user intention can be explicitly modeled for each user-item pair. We then introduce a dynamic transformer encoder (DTE) to capture user-specific intra-item relationships among item candidates by seamlessly accommodating the learnt latent user intentions via IDM. As such, RAISE is able to perform user-specific prediction without increasing the depth (number of blocks) and width (number of heads) of the prediction model. Empirical study on four public datasets shows the superiority of our proposed RAISE, with up to 13.95%, 12.30%, and 13.03% relative improvements evaluated by Precision, MAP, and NDCG respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

Attention over Self-attention:Intention-aware Re-ranking with Dynamic Transformer Encoders for Recommendation

Lin¹,

Zang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Domain Adaptation (DA) aims to train a target-domain classifier with samples from source and target domains (Lu et al 2015). When the labels of samples in the target domain are unavailable, DA is known as unsupervised DA (UDA) (Zhong et al 2020;Fang et al 2020), which has been applied to address diverse real-world problems, such as computer version (Zhang et al 2020c;Dong et al 2019Dong et al , 2020b, natural language processing (Lee and Jha 2019; Guo, Pasunuru, and Bansal 2020), and recommender system (Zhang et al 2017;Yu, Wang, and Yuan 2019;Lu et al 2020) Significant theoretical advances have been achieved in UDA. Pioneering theoretical work was proposed by Ben-David et al (2007).…”

Section: Introductionmentioning

confidence: 99%

How does the Combined Risk Affect the Performance of Unsupervised Domain Adaptation Approaches?

Zhong¹,

Fang²,

Liu³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Unsupervised domain adaptation (UDA) aims to train a target classifier with labeled samples from the source domain and unlabeled samples from the target domain. Classical UDA learning bounds show that target risk is upper bounded by three terms: source risk, distribution discrepancy, and combined risk. Based on the assumption that the combined risk is a small fixed value, methods based on this bound train a target classifier by only minimizing estimators of the source risk and the distribution discrepancy. However, the combined risk may increase when minimizing both estimators, which makes the target risk uncontrollable. Hence the target classifier cannot achieve ideal performance if we fail to control the combined risk. To control the combined risk, the key challenge takes root in the unavailability of the labeled samples in the target domain. To address this key challenge, we propose a method named E-MixNet. E-MixNet employs enhanced mixup, a generic vicinal distribution, on the labeled source samples and pseudo-labeled target samples to calculate a proxy of the combined risk. Experiments show that the proxy can effectively curb the increase of the combined risk when minimizing the source risk and distribution discrepancy. Furthermore, we show that if the proxy of the combined risk is added into loss functions of four representative UDA methods, their performance is also improved.

show abstract

“…Facial expression is one of the most powerful, natural, and universal symbols for human beings to show their emotions [34]. With the popularization of mobile devices and social networks [79], more and more people prefer to record their daily activities [66], share their feelings, and express their opinions through images and videos, which generates a massive quantity of multimedia data [21,39,72]. The primary task of the current extensively-applied facial recognition system [68] is to identify and verify people.…”

Section: Introductionmentioning

confidence: 99%

IExpressNet

Zhu

Luo

Zhao

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Existing methods on facial expression recognition (FER) are mainly trained in the setting when all expression classes are fixed in advance. However, in real applications, expression classes are becoming increasingly fine-grained and incremental. To deal with sequential expression classes, we can fine-tune or retrain these models, but this often results in poor performance or large computing resources consumption. To address these problems, we develop an Incremental Facial Expression Recognition Network (IExpress-Net), which can learn a competitive multi-class classifier at any time with a lower requirement of computing resources. Specifically, IExpressNet consists of two novel components. First, we construct an exemplar set by dynamically selecting representative samples from old expression classes. Then, the exemplar set and new expression classes samples constitute the training set. Second, we design a novel center-expression-distilled loss. As for facial expression in the wild, center-expression-distilled loss enhances the discriminative power of the deeply learned features and prevents catastrophic forgetting. Extensive experiments are conducted on two large-scale FER datasets in the wild, RAF-DB and AffectNet. The results demonstrate the superiority of the proposed method as compared to state-of-the-art incremental learning approaches. CCS CONCEPTS • Information systems → Multimedia information systems; • Human-centered computing → Human computer interaction (HCI).

show abstract

An Input-aware Factorization Machine for Sparse Prediction

Cited by 34 publications

References 13 publications

Attention over Self-attention:Intention-aware Re-ranking with Dynamic Transformer Encoders for Recommendation

Attention over Self-attention:Intention-aware Re-ranking with Dynamic Transformer Encoders for Recommendation

How does the Combined Risk Affect the Performance of Unsupervised Domain Adaptation Approaches?

IExpressNet

Contact Info

Product

Resources

About