Jie Xu

Junfan Chen, Richong Zhang, Yongyi Mao et al.

Few-shot text classification has recently been promoted by the meta-learning paradigm which aims to identify target classes with knowledge transferred from source classes with sets of small tasks named episodes. Despite their success, existing works building their meta-learner based on Prototypical Networks are unsatisfactory in learning discriminative text representations between similar classes, which may lead to contradictions during label prediction. In addition, the tasklevel and instance-level overfitting problems in few-shot text classification caused by a few training examples are not sufficiently tackled. In this work, we propose a contrastive learning framework named ContrastNet to tackle both discriminative representation and overfitting problems in few-shot text classification. ContrastNet learns to pull closer text representations belonging to the same class and push away text representations belonging to different classes, while simultaneously introducing unsupervised contrastive regularization at both task-level and instance-level to prevent overfitting. Experiments on 8 few-shot text classification datasets show that ContrastNet outperforms the current state-of-the-art models.

Jie Xu, Min Ding

Recent advancements in computational power and algorithms have enabled unabridged data (e.g., raw images or audio) to be used as input in some models (e.g., deep learning). However, the black box nature of such models reduces their likelihood of adoption by marketing scholars. Our paradigm of analysis, the Transparent Model of Unabridged Data (TMUD), enables researchers to investigate the inner workings of such black box models by incorporating an ex ante filtration module and an ex post experimentation module. We empirically demonstrate the TMUD by investigating the role of facial components and sexual dimorphism in face perceptions, which have implications for four marketing contexts: advertisement (perceptions of approachability, trustworthiness, and competence), brand (perceptions of whether a face represents a brand's typical customer), category (perceptions of whether a face represents a category's typical customer), and customer persona (perceptions of whether a face represents the persona of a brand's customer segment). Our results reveal new and useful findings that enrich the existing literature on face perception, most of which is based on abridged attributes (e.g., width of mouth). The TMUD has great potential to be a useful paradigm for generating theoretical insights and may encourage more marketing researchers and practitioners to use unabridged data.

Yuandong Wang, Hongzhi Yin, Tong Chen et al.

In recent years, ride-hailing services have been increasingly prevalent as they provide huge convenience for passengers. As a fundamental problem, the timely prediction of passenger demands in different regions is vital for effective traffic flow control and route planning. As both spatial and temporal patterns are indispensable passenger demand prediction, relevant research has evolved from pure time series to graph-structured data for modeling historical passenger demand data, where a snapshot graph is constructed for each time slot by connecting region nodes via different relational edges (e.g., origin-destination relationship, geographical distance, etc.). Consequently, the spatiotemporal passenger demand records naturally carry dynamic patterns in the constructed graphs, where the edges also encode important information about the directions and volume (i.e., weights) of passenger demands between two connected regions. However, existing graph-based solutions fail to simultaneously consider those three crucial aspects of dynamic, directed, and weighted (DDW) graphs, leading to limited expressiveness when learning graph representations for passenger demand prediction. Therefore, we propose a novel spatiotemporal graph attention network, namely Gallat (Graph prediction with all attention) as a solution. In Gallat, by comprehensively incorporating those three intrinsic properties of DDW graphs, we build three attention layers to fully capture the spatiotemporal dependencies among different regions across all historical time slots. Moreover, the model employs a subtask to conduct pretraining so that it can obtain accurate results more quickly. We evaluate the proposed model on real-world datasets, and our experimental results demonstrate that Gallat outperforms the state-of-the-art approaches.

Jie Xu, Wei Zhang, Fei Wang

As deep learning models are usually massive and complex, distributed learning is essential for increasing training efficiency. Moreover, in many real-world application scenarios like healthcare, distributed learning can also keep the data local and protect privacy. A popular distributed learning strategy is federated learning, where there is a central server storing the global model and a set of local computing nodes updating the model parameters with their corresponding data. The updated model parameters will be processed and transmitted to the central server, which leads to heavy communication costs. Recently, asynchronous decentralized distributed learning has been proposed and demonstrated to be a more efficient and practical strategy where there is no central server, so that each computing node only communicates with its neighbors. Although no raw data will be transmitted across different local nodes, there is still a risk of information leak during the communication process for malicious participants to make attacks. In this paper, we present a differentially private version of asynchronous decentralized parallel SGD (ADPSGD) framework, or A(DP)$^2$SGD for short, which maintains communication efficiency of ADPSGD and prevents the inference from malicious participants. Specifically, R{é}nyi differential privacy is used to provide tighter privacy analysis for our composite Gaussian mechanisms while the convergence rate is consistent with the non-private version. Theoretical analysis shows A(DP)$^2$SGD also converges at the optimal $\mathcal{O}(1/\sqrt{T})$ rate as SGD. Empirically, A(DP)$^2$SGD achieves comparable model accuracy as the differentially private version of Synchronous SGD (SSGD) but runs much faster than SSGD in heterogeneous computing environments.

James Dyer, Martin Dyer, Jie Xu

We present novel homomorphic encryption schemes for integer arithmetic, intended for use in secure single-party computation in the cloud. These schemes are capable of securely computing only low degree polynomials homomorphically, but this appears sufficient for most practical applications. In this setting, our schemes lead to practical key and ciphertext sizes. We present a sequence of generalisations of our basic schemes, with increasing levels of security, but decreasing practicality. We have evaluated the first four of these algorithms by computing a low-degree inner product. The timings of these computations are extremely favourable. Finally, we use our ideas to derive a fully homomorphic system, which appears impractical, but can homomorphically evaluate arbitrary Boolean circuits.