Zesheng Ye
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zesheng Ye has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Lina Yao
University of New South Wales, Sydney, Australia
6 shared publications - Jing Du
Department of Computer Science and Engineering The University of New South Wales (UNSW) Sydney Australia
4 shared publications - Bin Guo
Northwestern Polytechnical University
3 shared publications - Zhiwen Yu
Harbin Engineering University
3 shared publications - Haowei Lou
University of New South Wales, Sydney, NSW, Australia
1 shared publication - Sylvia Maria Gustin
NeuroRecovery Research Hub, School of Psychology The University of New South Wales (UNSW) Sydney Sydney New South Wales Australia
1 shared publication
Latest Publications
- 22 July 2024
- Published by World Scientific Pub Co Pte Ltd in Guidance, Navigation and Control
- Vol. 04 (03)
Transferrable Subject-Independent Feature Representation for Discriminating EEG-Based Brain Signals
Subject-independent Electroencephalography (EEG) recognition remains challenging due to inherent variability of brain anatomy across different subjects. Such variability is further complicated by the Volume Conduction Effect (VCE) that introduces channel-interference noise, exacerbating subject-specific biases in the recorded EEG signals. Existing studies, often relying large datasets and entangled spatial-temporal features, struggle to overcome this bias, particularly in scenarios with limited EEG data. To this end, we propose a Temporal-Connective EEG Representation Learning (TCERL) framework that disentangles temporal and spatial feature learning. TCERL first employs an one-dimensional convolutional network to extract channel-specific temporal features, mitigating channel-interference noise caused by VCE. Building upon these temporal features, TCERL then leverages Graph Neural Networks to extract subject-invariant topological features from a functional brain network, constructed using the channel-specific features as nodes and functional connectivity as the adjacency matrix. This approach allows TCERL to capture robust representations of brain activity patterns that generalize well across different subjects. Our empirical experiment demonstrates that TCERL outperforms state-of-the-art across a range of training subjects on four public benchmarks and is less sensitive to subject variability. The performance gain is highlighted when limited subjects are available, suggesting the robustness and transferability of the proposed method. Source code can be found in: https://github.com/haoweilou/TCERL .
- 22 July 2024
- Published by World Scientific Pub Co Pte Ltd in Guidance, Navigation and Control
- Vol. 04 (03)
Transferrable Subject-Independent Feature Representation for Discriminating EEG-Based Brain Signals
Subject-independent Electroencephalography (EEG) recognition remains challenging due to inherent variability of brain anatomy across different subjects. Such variability is further complicated by the Volume Conduction Effect (VCE) that introduces channel-interference noise, exacerbating subject-specific biases in the recorded EEG signals. Existing studies, often relying large datasets and entangled spatial-temporal features, struggle to overcome this bias, particularly in scenarios with limited EEG data. To this end, we propose a Temporal-Connective EEG Representation Learning (TCERL) framework that disentangles temporal and spatial feature learning. TCERL first employs an one-dimensional convolutional network to extract channel-specific temporal features, mitigating channel-interference noise caused by VCE. Building upon these temporal features, TCERL then leverages Graph Neural Networks to extract subject-invariant topological features from a functional brain network, constructed using the channel-specific features as nodes and functional connectivity as the adjacency matrix. This approach allows TCERL to capture robust representations of brain activity patterns that generalize well across different subjects. Our empirical experiment demonstrates that TCERL outperforms state-of-the-art across a range of training subjects on four public benchmarks and is less sensitive to subject variability. The performance gain is highlighted when limited subjects are available, suggesting the robustness and transferability of the proposed method. Source code can be found in: https://github.com/haoweilou/TCERL .
- Open Access
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2091-2101
Identifiability of Cross-Domain Recommendation via Causal Subspace Disentanglement
Cross-Domain Recommendation~(CDR) seeks to enable effective knowledge transfer across domains. Most existing works rely on either representation alignment or transformation bridges, but they come with shortcomings regarding identifiability of domain-shared and domain-specific latent factors. Specifically, while CDR describes user representations as a joint distribution over two domains, these methods fail to account for its joint identifiability as they primarily fixate on the marginal distribution within a particular domain. Such a failure may overlook the conditionality between two domains and how it contributes to latent factor disentanglement, leading to negative transfer when domains are weakly correlated. In this study, we explore what should and should not be transferred in cross-domain user representations from a causality perspective. We propose a Hierarchical causal subspace disentanglement approach to explore the Joint IDentifiability of cross-domain joint distribution, termed HJID, to preserve domain-specific behaviors from domain-shared factors. HJID abides by the feature hierarchy and divides user representations into generic shallow subspace and domain-oriented deep subspaces. We first encode the generic pattern in the shallow subspace by minimizing the Maximum Mean Discrepancy of initial layer activation. Then, to dissect how domain-oriented latent factors are encoded in deeper layers activation, we construct a cross-domain causality-based data generation graph, which identifies cross-domain consistent and domain-specific components, adhering to the Minimal Change principle. This allows HJID to maintain stability whilst discovering unique factors for different domains, all within a generative framework of invertible transformations that guarantee the joint identifiability. With experiments on real-world datasets, we show that HJID outperforms SOTA methods on both strong- and weak-correlation CDR tasks.
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2091-2101
Open AccessIdentifiability of Cross-Domain Recommendation via Causal Subspace Disentanglement
Cross-Domain Recommendation~(CDR) seeks to enable effective knowledge transfer across domains. Most existing works rely on either representation alignment or transformation bridges, but they come with shortcomings regarding identifiability of domain-shared and domain-specific latent factors. Specifically, while CDR describes user representations as a joint distribution over two domains, these methods fail to account for its joint identifiability as they primarily fixate on the marginal distribution within a particular domain. Such a failure may overlook the conditionality between two domains and how it contributes to latent factor disentanglement, leading to negative transfer when domains are weakly correlated. In this study, we explore what should and should not be transferred in cross-domain user representations from a causality perspective. We propose a Hierarchical causal subspace disentanglement approach to explore the Joint IDentifiability of cross-domain joint distribution, termed HJID, to preserve domain-specific behaviors from domain-shared factors. HJID abides by the feature hierarchy and divides user representations into generic shallow subspace and domain-oriented deep subspaces. We first encode the generic pattern in the shallow subspace by minimizing the Maximum Mean Discrepancy of initial layer activation. Then, to dissect how domain-oriented latent factors are encoded in deeper layers activation, we construct a cross-domain causality-based data generation graph, which identifies cross-domain consistent and domain-specific components, adhering to the Minimal Change principle. This allows HJID to maintain stability whilst discovering unique factors for different domains, all within a generative framework of invertible transformations that guarantee the joint identifiability. With experiments on real-world datasets, we show that HJID outperforms SOTA methods on both strong- and weak-correlation CDR tasks.
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5123-5127
NP-SSL: A Modular and Extensible Self-supervised Learning Library with Neural Processes
Neural Processes (NPs) are a family of supervised density estimators devoted to probabilistic function approximation with meta-learning. Despite extensive research on the subject, the absence of a unified framework for NPs leads to varied architectural solutions across diverse studies. This non-consensus poses challenges to reproducing and benchmarking different NPs. Moreover, existing codebases mainly prioritize generative density estimation, yet rarely consider expanding the capability of NPs to self-supervised representation learning, which however has gained growing importance in data mining applications. To this end, we present NP-SSL, a modular and configurable framework with built-in support, requiring minimal effort to 1) implement classical NPs architectures; 2) customize specific components; 3) integrate hybrid training scheme (e.g., contrastive); and 4) extend NPs to act as a self-supervised learning toolkit, producing latent representations of data, and facilitating diverse downstream predictive tasks. To illustrate, we discuss a case study that applies NP-SSL to model time-series data. We interpret that NP-SSL can handle different predictive tasks such as imputation and forecasting, by a simple switch in data samplings, without significant change to the underlying structure. We hope this study can reduce the workload of future research on leveraging NPs to tackle more a broader range of real-world data mining applications. Code and documentation are at https://github.com/zyecs/NP-SSL.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 5123-5127
Open AccessNP-SSL: A Modular and Extensible Self-supervised Learning Library with Neural Processes
Neural Processes (NPs) are a family of supervised density estimators devoted to probabilistic function approximation with meta-learning. Despite extensive research on the subject, the absence of a unified framework for NPs leads to varied architectural solutions across diverse studies. This non-consensus poses challenges to reproducing and benchmarking different NPs. Moreover, existing codebases mainly prioritize generative density estimation, yet rarely consider expanding the capability of NPs to self-supervised representation learning, which however has gained growing importance in data mining applications. To this end, we present NP-SSL, a modular and configurable framework with built-in support, requiring minimal effort to 1) implement classical NPs architectures; 2) customize specific components; 3) integrate hybrid training scheme (e.g., contrastive); and 4) extend NPs to act as a self-supervised learning toolkit, producing latent representations of data, and facilitating diverse downstream predictive tasks. To illustrate, we discuss a case study that applies NP-SSL to model time-series data. We interpret that NP-SSL can handle different predictive tasks such as imputation and forecasting, by a simple switch in data samplings, without significant change to the underlying structure. We hope this study can reduce the workload of future research on leveraging NPs to tackle more a broader range of real-world data mining applications. Code and documentation are at https://github.com/zyecs/NP-SSL.
- Open Access
- 30 August 2023
- Published by Elsevier in Pattern Recognition
- Vol. 145
Self-supervised cross-modal visual retrieval from brain activities
- 30 August 2023
- Published by Elsevier in Pattern Recognition
- Vol. 145
Open AccessSelf-supervised cross-modal visual retrieval from brain activities
- Open Access
- 27 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 481-489
IDNP: Interest Dynamics Modeling Using Generative Neural Processes for Sequential Recommendation
Recent sequential recommendation models rely increasingly on consecutive short-term user-item interaction sequences to model user interests. These approaches have, however, raised concerns about both short- and long-term interests. (1) short-term: interaction sequences may not result from a monolithic interest, but rather from several intertwined interests, even within a short period of time, resulting in their failures to model skip behaviors; (2) long-term: interaction sequences are primarily observed sparsely at discrete intervals, other than consecutively over the long run. This renders difficulty in inferring long-term interests, since only discrete interest representations can be derived, without taking into account interest dynamics across sequences. In this study, we address these concerns by learning (1) multi-scale representations of short-term interests; and (2) dynamics-aware representations of long-term interests. To this end, we present an Interest Dynamics modeling framework using generative Neural Processes, coined IDNP, to model user interests from a functional perspective. IDNP learns a global interest function family to define each user's long-term interest as a function instantiation, manifesting interest dynamics through function continuity. Specifically, IDNP first encodes each user's short-term interactions into multi-scale representations, which are then summarized as user context. By combining latent global interest with user context, IDNP then reconstructs long-term user interest functions and predicts interactions at upcoming query timestep. Moreover, IDNP can model such interest functions even when interaction sequences are limited and non-consecutive. Extensive experiments on four real-world datasets demonstrate that our model outperforms the state-of-the-art on various evaluation metrics.
- 27 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 481-489
Open AccessIDNP: Interest Dynamics Modeling Using Generative Neural Processes for Sequential Recommendation
Recent sequential recommendation models rely increasingly on consecutive short-term user-item interaction sequences to model user interests. These approaches have, however, raised concerns about both short- and long-term interests. (1) short-term: interaction sequences may not result from a monolithic interest, but rather from several intertwined interests, even within a short period of time, resulting in their failures to model skip behaviors; (2) long-term: interaction sequences are primarily observed sparsely at discrete intervals, other than consecutively over the long run. This renders difficulty in inferring long-term interests, since only discrete interest representations can be derived, without taking into account interest dynamics across sequences. In this study, we address these concerns by learning (1) multi-scale representations of short-term interests; and (2) dynamics-aware representations of long-term interests. To this end, we present an Interest Dynamics modeling framework using generative Neural Processes, coined IDNP, to model user interests from a functional perspective. IDNP learns a global interest function family to define each user's long-term interest as a function instantiation, manifesting interest dynamics through function continuity. Specifically, IDNP first encodes each user's short-term interactions into multi-scale representations, which are then summarized as user context. By combining latent global interest with user context, IDNP then reconstructs long-term user interest functions and predicts interactions at upcoming query timestep. Moreover, IDNP can model such interest functions even when interaction sequences are limited and non-consecutive. Extensive experiments on four real-world datasets demonstrate that our model outperforms the state-of-the-art on various evaluation metrics.
- Open Access
- 6 July 2022
- Published by Association for Computing Machinery (ACM)
- p. 1927-1932
Socially-aware Dual Contrastive Learning for Cold-Start Recommendation
Social recommendation with Graph Neural Networks(GNNs) learns to represent cold users by fusing user-user social relations with user-item interactions, thereby alleviating the cold-start problem associated with recommender systems. Despite being well adapted to social relations and user-item interactions, these supervised models are still susceptible to popularity bias. Contrastive learning helps resolve this dilemma by identifying the properties that distinguish positive from negative samples. In its previous combinations with recommender systems, social relationships and cold-start cases in this context are not considered. Also, they primarily focus on collaborative features between users and items, leaving the similarity between items under-utilized. In this work, we propose socially-aware dual contrastive learning for cold-start recommendation, where cold users can be modeled in the same way as warm users. To take full advantage of social relations, we create dynamic node embeddings for each user by aggregating information from different neighbors according to each different query item, in the form of user-item pairs. We further design a dual-branch self-supervised contrastive objective to account for user-item collaborative features and item-item mutual information, respectively. On one hand, our framework eliminates popularity bias with proper negative sampling in contrastive learning, without extra ground-truth supervision. On the other hand, we extend previous contrastive learning methods to provide a solution to cold-start problem with social relations included. Extensive experiments on two real-world social recommendation datasets demonstrate its effectiveness.
- 6 July 2022
- Published by Association for Computing Machinery (ACM)
- p. 1927-1932
Open AccessSocially-aware Dual Contrastive Learning for Cold-Start Recommendation
Social recommendation with Graph Neural Networks(GNNs) learns to represent cold users by fusing user-user social relations with user-item interactions, thereby alleviating the cold-start problem associated with recommender systems. Despite being well adapted to social relations and user-item interactions, these supervised models are still susceptible to popularity bias. Contrastive learning helps resolve this dilemma by identifying the properties that distinguish positive from negative samples. In its previous combinations with recommender systems, social relationships and cold-start cases in this context are not considered. Also, they primarily focus on collaborative features between users and items, leaving the similarity between items under-utilized. In this work, we propose socially-aware dual contrastive learning for cold-start recommendation, where cold users can be modeled in the same way as warm users. To take full advantage of social relations, we create dynamic node embeddings for each user by aggregating information from different neighbors according to each different query item, in the form of user-item pairs. We further design a dual-branch self-supervised contrastive objective to account for user-item collaborative features and item-item mutual information, respectively. On one hand, our framework eliminates popularity bias with proper negative sampling in contrastive learning, without extra ground-truth supervision. On the other hand, we extend previous contrastive learning methods to provide a solution to cold-start problem with social relations included. Extensive experiments on two real-world social recommendation datasets demonstrate its effectiveness.