Alert button
Picture for Hui Fang

Hui Fang

Alert button

CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning

Sep 16, 2023
Qingtian Bian, Jiaxing Xu, Hui Fang, Yiping Ke

Figure 1 for CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning
Figure 2 for CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning
Figure 3 for CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning
Figure 4 for CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning

The motivations of users to make interactions can be divided into static preference and dynamic interest. To accurately model user representations over time, recent studies in sequential recommendation utilize information propagation and evolution to mine from batches of arriving interactions. However, they ignore the fact that people are easily influenced by the recent actions of other users in the contextual scenario, and applying evolution across all historical interactions dilutes the importance of recent ones, thus failing to model the evolution of dynamic interest accurately. To address this issue, we propose a Context-Aware Pseudo-Multi-Task Recommender System (CPMR) to model the evolution in both historical and contextual scenarios by creating three representations for each user and item under different dynamics: static embedding, historical temporal states, and contextual temporal states. To dually improve the performance of temporal states evolution and incremental recommendation, we design a Pseudo-Multi-Task Learning (PMTL) paradigm by stacking the incremental single-target recommendations into one multi-target task for joint optimization. Within the PMTL paradigm, CPMR employs a shared-bottom network to conduct the evolution of temporal states across historical and contextual scenarios, as well as the fusion of them at the user-item level. In addition, CPMR incorporates one real tower for incremental predictions, and two pseudo towers dedicated to updating the respective temporal states based on new batches of interactions. Experimental results on four benchmark recommendation datasets show that CPMR consistently outperforms state-of-the-art baselines and achieves significant gains on three of them. The code is available at: https://github.com/DiMarzioBian/CPMR.

* ACM International Conference on Information and Knowledge Management(CIKM '23), October 21-25,2023,Birmingham,United Kingdom  
* Accepted by CIKM 2023. Alias: "Modeling Context-Aware Temporal Dynamics via Pseudo-Multi-Task Learning" 
Viaarxiv icon

Zero-shot Query Reformulation for Conversational Search

Jul 18, 2023
Dayu Yang, Yue Zhang, Hui Fang

As the popularity of voice assistants continues to surge, conversational search has gained increased attention in Information Retrieval. However, data sparsity issues in conversational search significantly hinder the progress of supervised conversational search methods. Consequently, researchers are focusing more on zero-shot conversational search approaches. Nevertheless, existing zero-shot methods face three primary limitations: they are not universally applicable to all retrievers, their effectiveness lacks sufficient explainability, and they struggle to resolve common conversational ambiguities caused by omission. To address these limitations, we introduce a novel Zero-shot Query Reformulation (ZeQR) framework that reformulates queries based on previous dialogue contexts without requiring supervision from conversational search data. Specifically, our framework utilizes language models designed for machine reading comprehension tasks to explicitly resolve two common ambiguities: coreference and omission, in raw queries. In comparison to existing zero-shot methods, our approach is universally applicable to any retriever without additional adaptation or indexing. It also provides greater explainability and effectively enhances query intent understanding because ambiguities are explicitly and proactively resolved. Through extensive experiments on four TREC conversational datasets, we demonstrate the effectiveness of our method, which consistently outperforms state-of-the-art baselines.

* Accepted by ICTIR 2023 
Viaarxiv icon

An Exploration Study of Mixed-initiative Query Reformulation in Conversational Passage Retrieval

Jul 17, 2023
Dayu Yang, Yue Zhang, Hui Fang

Figure 1 for An Exploration Study of Mixed-initiative Query Reformulation in Conversational Passage Retrieval
Figure 2 for An Exploration Study of Mixed-initiative Query Reformulation in Conversational Passage Retrieval
Figure 3 for An Exploration Study of Mixed-initiative Query Reformulation in Conversational Passage Retrieval
Figure 4 for An Exploration Study of Mixed-initiative Query Reformulation in Conversational Passage Retrieval

In this paper, we report our methods and experiments for the TREC Conversational Assistance Track (CAsT) 2022. In this work, we aim to reproduce multi-stage retrieval pipelines and explore one of the potential benefits of involving mixed-initiative interaction in conversational passage retrieval scenarios: reformulating raw queries. Before the first ranking stage of a multi-stage retrieval pipeline, we propose a mixed-initiative query reformulation module, which achieves query reformulation based on the mixed-initiative interaction between the users and the system, as the replacement for the neural reformulation method. Specifically, we design an algorithm to generate appropriate questions related to the ambiguities in raw queries, and another algorithm to reformulate raw queries by parsing users' feedback and incorporating it into the raw query. For the first ranking stage of our multi-stage pipelines, we adopt a sparse ranking function: BM25, and a dense retrieval method: TCT-ColBERT. For the second-ranking step, we adopt a pointwise reranker: MonoT5, and a pairwise reranker: DuoT5. Experiments on both TREC CAsT 2021 and TREC CAsT 2022 datasets show the effectiveness of our mixed-initiative-based query reformulation method on improving retrieval performance compared with two popular reformulators: a neural reformulator: CANARD-T5 and a rule-based reformulator: historical query reformulator(HQE).

* The Thirty-First Text REtrieval Conference (TREC 2022) Proceedings 
Viaarxiv icon

Contactless hand tremor amplitude measurement using smartphones: development and pilot evaluation

Apr 28, 2023
James Bungay, Osasenaga Emokpae, Samuel D. Relton, Jane Alty, Stefan Williams, Hui Fang, David C. Wong

Figure 1 for Contactless hand tremor amplitude measurement using smartphones: development and pilot evaluation
Figure 2 for Contactless hand tremor amplitude measurement using smartphones: development and pilot evaluation
Figure 3 for Contactless hand tremor amplitude measurement using smartphones: development and pilot evaluation
Figure 4 for Contactless hand tremor amplitude measurement using smartphones: development and pilot evaluation

Background: Physiological tremor is defined as an involuntary and rhythmic shaking. Tremor of the hand is a key symptom of multiple neurological diseases, and its frequency and amplitude differs according to both disease type and disease progression. In routine clinical practice, tremor frequency and amplitude are assessed by expert rating using a 0 to 4 integer scale. Such ratings are subjective and have poor inter-rater reliability. There is thus a clinical need for a practical and accurate method for objectively assessing hand tremor. Objective: to develop a proof of principle method to measure hand tremor amplitude from smartphone videos. Methods: We created a computer vision pipeline that automatically extracts salient points on the hand and produces a 1-D time series of movement due to tremor, in pixels. Using the smartphones' depth measurement, we convert this measure into real distance units. We assessed the accuracy of the method using 60 videos of simulated tremor of different amplitudes from two healthy adults. Videos were taken at distances of 50, 75 and 100 cm between hand and camera. The participants had skin tone II and VI on the Fitzpatrick scale. We compared our method to a gold-standard measurement from a slide rule. Bland-Altman methods agreement analysis indicated a bias of 0.04 cm and 95% limits of agreement from -1.27 to 1.20 cm. Furthermore, we qualitatively observed that the method was robust to differences in skin tone and limited occlusion, such as a band-aid affixed to the participant's hand. Clinical relevance: We have demonstrated how tremor amplitude can be measured from smartphone videos. In conjunction with tremor frequency, this approach could be used to help diagnose and monitor neurological diseases

* Accepted to IEEE EMBC 2023, Sydney (pre-refereed version) 
Viaarxiv icon

Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors

Mar 14, 2023
Moritz Neun, Christian Eichenberger, Henry Martin, Markus Spanring, Rahul Siripurapu, Daniel Springer, Leyan Deng, Chenwang Wu, Defu Lian, Min Zhou, Martin Lumiste, Andrei Ilie, Xinhua Wu, Cheng Lyu, Qing-Long Lu, Vishal Mahajan, Yichao Lu, Jiezhang Li, Junjun Li, Yue-Jiao Gong, Florian Grötschla, Joël Mathys, Ye Wei, He Haitao, Hui Fang, Kevin Malm, Fei Tang, Michael Kopp, David Kreil, Sepp Hochreiter

Figure 1 for Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors
Figure 2 for Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors
Figure 3 for Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors
Figure 4 for Traffic4cast at NeurIPS 2022 -- Predict Dynamics along Graph Edges from Sparse Node Data: Whole City Traffic and ETA from Stationary Vehicle Detectors

The global trends of urbanization and increased personal mobility force us to rethink the way we live and use urban space. The Traffic4cast competition series tackles this problem in a data-driven way, advancing the latest methods in machine learning for modeling complex spatial systems over time. In this edition, our dynamic road graph data combine information from road maps, $10^{12}$ probe data points, and stationary vehicle detectors in three cities over the span of two years. While stationary vehicle detectors are the most accurate way to capture traffic volume, they are only available in few locations. Traffic4cast 2022 explores models that have the ability to generalize loosely related temporal vertex data on just a few nodes to predict dynamic future traffic states on the edges of the entire road graph. In the core challenge, participants are invited to predict the likelihoods of three congestion classes derived from the speed levels in the GPS data for the entire road graph in three cities 15 min into the future. We only provide vehicle count data from spatially sparse stationary vehicle detectors in these three cities as model input for this task. The data are aggregated in 15 min time bins for one hour prior to the prediction time. For the extended challenge, participants are tasked to predict the average travel times on super-segments 15 min into the future - super-segments are longer sequences of road segments in the graph. The competition results provide an important advance in the prediction of complex city-wide traffic states just from publicly available sparse vehicle data and without the need for large amounts of real-time floating vehicle data.

* Pre-print under review, submitted to Proceedings of Machine Learning Research 
Viaarxiv icon

Decentralized Matrix Factorization with Heterogeneous Differential Privacy

Dec 01, 2022
Wentao Hu, Hui Fang

Figure 1 for Decentralized Matrix Factorization with Heterogeneous Differential Privacy
Figure 2 for Decentralized Matrix Factorization with Heterogeneous Differential Privacy
Figure 3 for Decentralized Matrix Factorization with Heterogeneous Differential Privacy
Figure 4 for Decentralized Matrix Factorization with Heterogeneous Differential Privacy

Conventional matrix factorization relies on centralized collection of users' data for recommendation, which might introduce an increased risk of privacy leakage especially when the recommender is untrusted. Existing differentially private matrix factorization methods either assume the recommender is trusted, or can only provide a uniform level of privacy protection for all users and items with untrusted recommender. In this paper, we propose a novel Heterogeneous Differentially Private Matrix Factorization algorithm (denoted as HDPMF) for untrusted recommender. To the best of our knowledge, we are the first to achieve heterogeneous differential privacy for decentralized matrix factorization in untrusted recommender scenario. Specifically, our framework uses modified stretching mechanism with an innovative rescaling scheme to achieve better trade off between privacy and accuracy. Meanwhile, by allocating privacy budget properly, we can capture homogeneous privacy preference within a user/item but heterogeneous privacy preference across different users/items. Theoretical analysis confirms that HDPMF renders rigorous privacy guarantee, and exhaustive experiments demonstrate its superiority especially in strong privacy guarantee, high dimension model and sparse dataset scenario.

Viaarxiv icon

Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring

Nov 14, 2022
Wenyuan Yang, Shuo Shao, Yue Yang, Xiyao Liu, Zhihua Xia, Gerald Schaefer, Hui Fang

Figure 1 for Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring
Figure 2 for Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring
Figure 3 for Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring
Figure 4 for Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring

Federated learning (FL) allows multiple participants to collaboratively build deep learning (DL) models without directly sharing data. Consequently, the issue of copyright protection in FL becomes important since unreliable participants may gain access to the jointly trained model. Application of homomorphic encryption (HE) in secure FL framework prevents the central server from accessing plaintext models. Thus, it is no longer feasible to embed the watermark at the central server using existing watermarking schemes. In this paper, we propose a novel client-side FL watermarking scheme to tackle the copyright protection issue in secure FL with HE. To our best knowledge, it is the first scheme to embed the watermark to models under the Secure FL environment. We design a black-box watermarking scheme based on client-side backdooring to embed a pre-designed trigger set into an FL model by a gradient-enhanced embedding method. Additionally, we propose a trigger set construction mechanism to ensure the watermark cannot be forged. Experimental results demonstrate that our proposed scheme delivers outstanding protection performance and robustness against various watermark removal attacks and ambiguity attack.

Viaarxiv icon

Understanding Diversity in Session-Based Recommendation

Aug 29, 2022
Qing Yin, Hui Fang, Zhu Sun, Yew-Soon Ong

Figure 1 for Understanding Diversity in Session-Based Recommendation
Figure 2 for Understanding Diversity in Session-Based Recommendation
Figure 3 for Understanding Diversity in Session-Based Recommendation
Figure 4 for Understanding Diversity in Session-Based Recommendation

Current session-based recommender systems (SBRSs) mainly focus on maximizing recommendation accuracy, while few studies have been devoted to improve diversity beyond accuracy. Meanwhile, it is unclear how the accuracy-oriented SBRSs perform in terms of diversity. Besides, the asserted "trade-off" relationship between accuracy and diversity has been increasingly questioned in the literature. Towards the aforementioned issues, we conduct a holistic study to particularly examine the recommendation performance of representative SBRSs w.r.t. both accuracy and diversity, striving for better understanding the diversity-related issues for SBRSs and providing guidance on designing diversified SBRSs. Particularly, for a fair and thorough comparison, we deliberately select state-of-the-art non-neural, deep neural, and diversified SBRSs, by covering more scenarios with appropriate experimental setups, e.g., representative datasets, evaluation metrics, and hyper-parameter optimization technique. Our empirical results unveil that: 1) non-diversified methods can also obtain satisfying performance on diversity, which might even surpass diversified ones; and 2) the relationship between accuracy and diversity is quite complex. Besides the "trade-off" relationship, they might be positively correlated with each other, that is, having a same-trend (win-win or lose-lose) relationship, which varies across different methods and datasets. Additionally, we further identify three possible influential factors on diversity in SBRSs (i.e., granularity of item categorization, session diversity of datasets, and length of recommendation lists).

Viaarxiv icon