High-level preferences as positive examples in contrastive learning for multi-interest sequential recommendation

Abstract

The sequential recommendation task based on the multi-interest framework aims to model multiple interests of users from different aspects to predict their future interactions. However, researchers rarely consider the differences in features between the interests generated by the model. In extreme cases, all interest capsules have the same meaning, leading to the failure of modeling users with multiple interests. To address this issue, we propose the High-level Preferences as positive examples in Contrastive Learning for multi-interest Sequence Recommendation framework (HPCL4SR), which uses contrastive learning to distinguish differences in interests based on user item interaction information. In order to find high-quality comparative examples, this paper introduces the category information to construct a global graph, learning the association between categories for high-level preference interest of users. Then, a multi-layer perceptron is used to adaptively fuse the low-level preference interest features of the user’s items and the high-level preference interest features of the categories. Finally, user multi-interest contrastive samples are obtained through item sequence information and corresponding categories, which are fed into contrastive learning to optimize model parameters and generate multi-interest representations that are more in line with the user sequence. In addition, when modeling the user’s item sequence information, in order to increase the differentiation between item representations, the category of the item is used to supervise the learning process. Extensive experiments on three real datasets demonstrate that our method outperforms existing multi-interest recommendation models.

Efficient feature redundancy reduction for image denoising

Abstract

It is challenging to deploy convolutional neural networks (CNNs) for image denoising on low-power devices which can suffer from computational and memory constraints. To address this limitation, a simple yet effective and efficient feature redundancy reduction-based network (FRRN) is proposed in this paper, which integrates a feature refinement block (FRB), an attention fusion block (AFB), and an enhancement block (EB). Specifically, the FRB distills structural information via two parallel sub-networks, selecting representative feature representations while suppressing spatial-channel redundancy. The AFB absorbs an attentive fusion mechanism to facilitate diverse features extracted from two sub-networks, emphasizing texture and structure details but alleviating harmful features from problematic regions. The subsequent EB further boosts the feature representation abilities. Aiming to enhance denoising performance at both pixel level and semantic level, a multi-loss scheme comprising three popular loss functions is leveraged to improve the robustness of the denoiser. Comprehensive quantitative and qualitative analyses demonstrate the superiority of the proposed FRRN.

Group-to-group recommendation with neural graph matching

Abstract

Nowadays, with the development of recommender systems, an emerging recommendation scenario called group-to-group recommendation has played a vital role in information acquisition for users. The new recommendation scenario seeks to recommend a group of related items to users with similar interests. To some extent, it alleviates the problem of point-to-point recommendations getting trapped in an information cocoon due to an over-reliance on user behaviors. For the new recommendation scenario, the existing recommendation methods cannot model the complex interactions between user groups and item groups, thus affecting the accuracy of the group-to-group recommendation. In this paper, we propose a group-to-group recommendation method, which abstracts user groups and item groups into graphs and calculates the similarity between two graphs based on graph matching, dubbed as GMRec. Specifically, we construct the graph of user groups and item groups and then calculate the graph similarity scores between user groups and item groups from two perspectives of feature matching and structure matching. Experimental results show that our model achieves higher accuracy than state-of-the-art models on three industrial datasets with different group sizes, with a maximum improvement of 8.2%.

Efficiently estimating node influence through group sampling over large graphs

Abstract

The huge amount of graph data necessitates sampling methods to support graph-based analysis applications. Node influence is to count the influential nodes with a given node in large graphs that has wide applications including product promotion and information diffusion in social networks. However, existing sampling methods mainly consider node degree to compute the node influence while ignoring the important connections in terms of groups in which nodes participate, resulting in inaccuracy of influence estimations. To this end, this paper proposes group sampling, called GVRW, to count the groups along with node degrees to evaluate node influence in large graphs. Specifically, GVRW changes the way of random walker traversing a large graph from one node to a random neighbor node of the groups to enlarge the sampling space for the sake of characterizing the nodes and groups simultaneously. Furthermore, we carefully design the corresponding estimated method to employ the samples to estimate the specific distributions of groups and node degrees to compute the node influence. Experimental results on real-world graph datasets show that our proposed sampling and estimating methods can accurately obtain the properties and approximate the node influences closer to the real values than existing methods.

A relation-aware representation approach for the question matching system

Abstract

Online question matching is the process of comparing user queries with system questions to find appropriate answers. This task has become increasingly important with the popularity of knowledge sharing social networks in product search and intelligent Q &A in customer service. Many previous studies have focused on designing complex semantic structures through the questions themselves. In fact, the online user’s queries accumulate a large number of similar sentences, which have been grouped by semantics in the retrieval system. However, how to use these sentences to enhance the understanding of system questions is rarely studied. In this paper, we propose a novel Relation-aware Semantic Enhancement Network (RSEN) model. Specifically, we leverage the labels of the history records to identify different semantically related sentences. Then, we construct an expanded relation network to integrate the representation of different semantic relations. Furthermore, we interact we integrate the features of the system question with the semantically related sentences to augment the semantic information. Finally, we evaluate our proposed RSEN on two publicly available datasets. The results demonstrate the effectiveness of our proposed RSEN method compared to the advanced baselines.

A reinforcement learning-based approach to testing GUI of moblie applications

Abstract

With the popularity of mobile devices, the software market of mobile applications has been booming in recent years. Android applications occupy a vast market share. However, the applications inevitably contain defects. Defects may affect the user experience and even cause severe economic losses. This paper proposes ATAC and ATPPO, which apply reinforcement learning to Android GUI testing to mitigate the state explosion problem. The article designs a new reward function and a new state representation. It also constructs two GUI testing models (ATAC and ATPPO) based on A2C and PPO algorithms to save memory space and accelerate training speed. Empirical studies on twenty open-source applications from GitHub demonstrate that: (1) ATAC performs best in 16 of 20 apps in code coverage and defects more exceptions; (2) ATPPO can get higher code coverage in 15 of 20 apps and defects more exceptions; (3) Compared with state-of-art tools Monkey and ARES, ATAC, and ATPPO shows higher code coverage and detects more errors. ATAC and ATPPO can not only cover more code coverage but also can effectively detect more exceptions. This paper also introduces Finite-State Machine into the reinforcement learning framework to avoid falling into the local optimal state, which provides high-level guidance for further improving the test efficiency.

Sampling hypergraphs via joint unbiased random walk

Abstract

Hypergraphs are instrumental in modeling complex relational systems that encompass a wide spectrum of high-order interactions among components. One prevalent analysis task is the properties estimation of large-scale hypergraphs, which involves selecting a subset of nodes and hyperedges while preserving the characteristics of the entire hypergraph. This paper aims to sample hypergraphs via random walks and is the first to perform unbiased random walks for sampling of nodes and hyperedges simultaneously in large-scale hypergraphs to the best of our knowledge. Initially, we analyze the stationary distributions of nodes and hyperedges for the simple random walk, and show that there is a high bias in both nodes and hyperedges. Subsequently, to eliminate the high bias of the simple random walk, we propose unbiased random walk strategies for nodes and hyperedges, respectively. Finally, a single joint walk schema is developed for sampling nodes and hyperedges simultaneously. To accelerate the convergence process, we employ delayed acceptance and history-aware techniques to assist our algorithm in achieving fast convergence. Extensive experimental results validate our theoretical findings, and the unbiased sampling algorithms for nodes and hyperedges have their complex hypergraph scenarios for which they are applicable. The joint random walk algorithm balanced the sampling applicable to both nodes and hyperedges.

Modeling dynamic spatiotemporal user preference for location prediction: a mutually enhanced method

Abstract

As the cornerstone of location-based services, location prediction aims to predict user’s next location through modeling user’s personal preference or travel sequential pattern. However, most existing methods only consider one of them and extremely sparse data makes it difficult to dynamically and comprehensively characterize user preference. In this paper, we propose a novel Dynamic Spatiotemporal User Preference (DSUP) model to characterize dynamic spatiotemporal user preference and integrate it with user’s travel sequential pattern for location prediction. Specifically, we design an interaction-aware graph attention network to learn the embeddings of locations and timeslots, and infer dynamic spatiotemporal user preference from the history travel locations and timeslots. Then, we combine user’s current travel preference with the impact of history travel sequential pattern to predict user’s next location. In addition, we predict user’s next travel timeslot and combine it with the temporal pattern of locations to enhance the location and timeslot prediction results mutually. We conduct extensive experiments on two public datasets Gowalla, Foursquare and our own Private Car dataset. The results on three datasets show that our method improves the accuracy and mean reciprocal rank of location prediction by 3%-11% and 7%-10% respectively.

LAMEE: a light all-MLP framework for time series prediction empowering recommendations

Abstract

Exogenous variables, unrelated to the recommendation system itself, can significantly enhance its performance. Therefore, integrating these time-evolving exogenous variables into a time series and conducting time series predictions can maximize the potential of recommendation systems. We refer to this task as Time Series Prediction Empowering Recommendations (TSPER). However, as a subtask within the recommendation system, TSPER faces unique challenges such as computational and data constraints, system evolution, and the need for performance and interpretability. To meet these unique needs, we propose a lightweight Multi-Layer Perceptron architecture with joint Time-Frequency information, named Light All-MLP with joint TimE-frEquency information (LAMEE). LAMEE utilizes a lightweight MLP architecture to achieve computing efficiency and adaptive online learning. Moreover, various strategies have been employed to improve the model, ensuring stable performance and model interpretability. Across multiple time series datasets potentially related to recommendation systems, LAMEE balances performance, efficiency, and interpretability, overall surpassing existing complex methods.

EMPNet: An extract-map-predict neural network architecture for cross-domain recommendation

Abstract

Cross-domain recommendation leverages a user’s historical interactions in the auxiliary domain to suggest items within the target domain, particularly for cold-start users with no prior activity in the target domain. Existing cross-domain recommendation models often overlook key aspects such as the complexities of transferring user interests between domains and the biases inherent in user behavior patterns. In contrast, our Extract-Map-Predict Neural Network Architecture (EMPNet) employs a disentanglement approach to map fine-grained user interests and utilize the biases inherent in the cross-domain recommendation. In feature extraction, we use the Bidirectional Encoder Representations from Transformers (BERT) and Identity-Enhanced Multi-Head Attention Mechanism to obtain the user and item feature vectors. In cross-domain user mapping, we disentangle the user feature vector into domain-shared and domain-specific interests for fine-grained cross-domain mapping to obtain the feature vector of cold-start users in the target domain. In rating prediction, we design a biased Attentional Factorization Machine (AFM) to utilize biases extracted from user and item features. We experimentally evaluate EMPNet on the Amazon dataset. The results show that it clearly outperforms the selected baselines.