arXiv subjects

Wei Wei

Publications and source records attributed to Wei Wei.

At least 271 records · Page 15Linked to original sources

arXiv2023

TCGF: A unified tensorized consensus graph framework for multi-view representation learning

Multi-view learning techniques have recently gained significant attention in the machine learning domain for their ability to leverage consistency and complementary information across multiple views. However, there remains a lack of sufficient research on generalized multi-view frameworks that unify existing works into a scalable and robust learning framework, as most current works focus on specific styles of multi-view models. Additionally, most multi-view learning works rely heavily on specific-scale scenarios and fail to effectively comprehend multiple scales holistically. These limitations hinder the effective fusion of essential information from multiple views, resulting in poor generalization. To address these limitations, this paper proposes a universal multi-view representation learning framework named Tensorized Consensus Graph Framework (TCGF). Specifically, it first provides a unified framework for existing multi-view works to exploit the representations for individual view, which aims to be suitable for arbitrary assumptions and different-scales datasets. Then, stacks them into a tensor under alignment basics as a high-order representation, allowing for the smooth propagation of consistency and complementary information across all views. Moreover, TCGF proposes learning a consensus embedding shared by adaptively collaborating all views to uncover the essential structure of the multi-view data, which utilizes view-consensus grouping effect to regularize the view-consensus representation. To further facilitate related research, we provide a specific implementation of TCGF for large-scale datasets, which can be efficiently solved by applying the alternating optimization strategy. Experimental results conducted on seven different-scales datasets indicate the superiority of the proposed TCGF against existing state-of-the-art multi-view learning methods.

cs.LG↗

arXiv2023

Most Probable Transitions from Metastable to Oscillatory Regimes in a Carbon Cycle System

Global climate changes are related to the ocean's store of carbon. We study a carbonate system of the upper ocean, which has metastable and oscillatory regimes, under small random fluctuations. We calculate the most probable transition path via a geometric minimum action method in the context of the large deviations theory. By examining the most probable transition paths from metastable to oscillatory regimes for various external carbon input rates, we find two different transition patterns, which gives us an early warning sign for the dramatic change in the carbonate state of the ocean.

math.NA↗

arXiv2023

An Optimal Control Method to Compute the Most Likely Transition Path for Stochastic Dynamical Systems with Jumps

Many complex real world phenomena exhibit abrupt, intermittent or jumping behaviors, which are more suitable to be described by stochastic differential equations under non-Gaussian Lévy noise. Among these complex phenomena, the most likely transition paths between metastable states are important since these rare events may have a high impact in certain scenarios. Based on the large deviation principle, the most likely transition path could be treated as the minimizer of the rate function upon paths that connect two points. One of the challenges to calculate the most likely transition path for stochastic dynamical systems under non-Gaussian Lévy noise is that the associated rate function can not be explicitly expressed by paths. For this reason, we formulate an optimal control problem to obtain the optimal state as the most likely transition path. We then develop a neural network method to solve this issue. Several experiments are investigated for both Gaussian and non-Gaussian cases.

math.NA↗

arXiv2023

Efficient Link Prediction in Continuous-Time Dynamic Networks using Optimal Transmission and Metropolis Hastings Sampling

Efficient link prediction in continuous-time dynamic networks is a challenging problem that has attracted much research attention in recent years. A widely used approach to dynamic network link prediction is to extract the local structure of the target link through temporal random walk on the network and learn node features using a coding model. However, this approach often assumes that candidate temporal neighbors follow some certain types of distributions, which may be inappropriate for real-world networks, thereby incurring information loss. To address this limitation, we propose a framework in continuous-time dynamic networks based on Optimal Transmission (OT) and Metropolis Hastings (MH) sampling (COM). Specifically, we use optimal transmission theory to calculate the Wasserstein distance between the current node and the time-valid candidate neighbors to minimize information loss in node information propagation. Additionally, we employ the MH algorithm to obtain higher-order structural relationships in the vicinity of the target link, as it is a Markov Chain Monte Carlo method and can flexibly simulate target distributions with complex patterns. We demonstrate the effectiveness of our proposed method through experiments on eight datasets from different fields.

math.DS↗

arXiv2023

An Empirical Study on the Language Modal in Visual Question Answering

Generalization beyond in-domain experience to out-of-distribution data is of paramount significance in the AI domain. Of late, state-of-the-art Visual Question Answering (VQA) models have shown impressive performance on in-domain data, partially due to the language priors bias which, however, hinders the generalization ability in practice. This paper attempts to provide new insights into the influence of language modality on VQA performance from an empirical study perspective. To achieve this, we conducted a series of experiments on six models. The results of these experiments revealed that, 1) apart from prior bias caused by question types, there is a notable influence of postfix-related bias in inducing biases, and 2) training VQA models with word-sequence-related variant questions demonstrated improved performance on the out-of-distribution benchmark, and the LXMERT even achieved a 10-point gain without adopting any debiasing methods. We delved into the underlying reasons behind these experimental results and put forward some simple proposals to reduce the models' dependency on language priors. The experimental results demonstrated the effectiveness of our proposed method in improving performance on the out-of-distribution benchmark, VQA-CPv2. We hope this study can inspire novel insights for future research on designing bias-reduction approaches.

cs.AI↗

arXiv2023

HyperBandit: Contextual Bandit with Hypernewtork for Time-Varying User Preferences in Streaming Recommendation

In real-world streaming recommender systems, user preferences often dynamically change over time (e.g., a user may have different preferences during weekdays and weekends). Existing bandit-based streaming recommendation models only consider time as a timestamp, without explicitly modeling the relationship between time variables and time-varying user preferences. This leads to recommendation models that cannot quickly adapt to dynamic scenarios. To address this issue, we propose a contextual bandit approach using hypernetwork, called HyperBandit, which takes time features as input and dynamically adjusts the recommendation model for time-varying user preferences. Specifically, HyperBandit maintains a neural network capable of generating the parameters for estimating time-varying rewards, taking into account the correlation between time features and user preferences. Using the estimated time-varying rewards, a bandit policy is employed to make online recommendations by learning the latent item contexts. To meet the real-time requirements in streaming recommendation scenarios, we have verified the existence of a low-rank structure in the parameter matrix and utilize low-rank factorization for efficient training. Theoretically, we demonstrate a sublinear regret upper bound against the best policy. Extensive experiments on real-world datasets show that the proposed HyperBandit consistently outperforms the state-of-the-art baselines in terms of accumulated rewards.

cs.IR↗

arXiv2023

Towards Autocomplete Strategies for Visualization Construction

Constructive visualization uses physical data units - tokens - to enable non-experts to create personalized visualizations engagingly. However, its physical nature limits efficiency and scalability. One potential solution to address this issue is autocomplete. By providing automated suggestions while still allowing for manual intervention, autocomplete can expedite visualization construction while maintaining expressivity. We conduct a speculative design study to examine how people would like to interact with a visualization authoring system that supports autocomplete. Our study identifies three types of autocomplete strategies and gains insights for designing future visualization authoring tools with autocomplete functionality. A free copy of this paper and all supplemental materials are available on our online repository https://osf.io/nu4z3/?view_only=594baee54d114a99ab381886fb32a126

cs.HC↗

arXiv2023

Observation of zero resistance above 100$^\circ$ K in Pb$_{10-x}$Cu$_x$(PO$_4$)$_6$O

Room-temperature superconductivity has always been regarded as the ultimate goal in the fields of solid-state physics and materials science, with its realization holding revolutionary significance, capable of triggering significant changes in energy transmission and storage. However, achieving it poses various challenges. Recent research revealed that material Pb$_{10-x}$Cu$_x$(PO$_4$)$_6$O displays room-temperature superconductivity under atmospheric pressure, sparking global interest in further exploration. Here, we utilized solid-phase synthesis to obtain a polycrystalline sample of Pb$_{10-x}$Cu$_x$(PO$_4$)$_6$O. X-ray diffraction confirmed its structural consistency with referenced literature. Zero resistance, which is important evidence for superconductivity, was observed above 100$^\circ$ K under ambient pressure in our experiment. Our finding indicates that Pb$_{10-x}$Cu$_x$(PO$_4$)$_6$O is a possible candidate for searching high-temperature superconductors.

cond-mat.supr-con↗

arXiv2023

Star Cluster Classification using Deep Transfer Learning with PHANGS-HST

Currently available star cluster catalogues from HST imaging of nearby galaxies heavily rely on visual inspection and classification of candidate clusters. The time-consuming nature of this process has limited the production of reliable catalogues and thus also post-observation analysis. To address this problem, deep transfer learning has recently been used to create neural network models which accurately classify star cluster morphologies at production scale for nearby spiral galaxies (D < 20 Mpc). Here, we use HST UV-optical imaging of over 20,000 sources in 23 galaxies from the Physics at High Angular Resolution in Nearby GalaxieS (PHANGS) survey to train and evaluate two new sets of models: i) distance-dependent models, based on cluster candidates binned by galaxy distance (9-12 Mpc, 14-18 Mpc, 18-24 Mpc), and ii) distance-independent models, based on the combined sample of candidates from all galaxies. We find that the overall accuracy of both sets of models is comparable to previous automated star cluster classification studies (~60-80 per cent) and show improvement by a factor of two in classifying asymmetric and multi-peaked clusters from PHANGS-HST. Somewhat surprisingly, while we observe a weak negative correlation between model accuracy and galactic distance, we find that training separate models for the three distance bins does not significantly improve classification accuracy. We also evaluate model accuracy as a function of cluster properties such as brightness, colour, and SED-fit age. Based on the success of these experiments, our models will provide classifications for the full set of PHANGS-HST candidate clusters (N ~ 200,000) for public release.

astro-ph.GA↗

arXiv2023

Towards Hierarchical Policy Learning for Conversational Recommendation with Hypergraph-based Reinforcement Learning

Conversational recommendation systems (CRS) aim to timely and proactively acquire user dynamic preferred attributes through conversations for item recommendation. In each turn of CRS, there naturally have two decision-making processes with different roles that influence each other: 1) director, which is to select the follow-up option (i.e., ask or recommend) that is more effective for reducing the action space and acquiring user preferences; and 2) actor, which is to accordingly choose primitive actions (i.e., asked attribute or recommended item) that satisfy user preferences and give feedback to estimate the effectiveness of the director's option. However, existing methods heavily rely on a unified decision-making module or heuristic rules, while neglecting to distinguish the roles of different decision procedures, as well as the mutual influences between them. To address this, we propose a novel Director-Actor Hierarchical Conversational Recommender (DAHCR), where the director selects the most effective option, followed by the actor accordingly choosing primitive actions that satisfy user preferences. Specifically, we develop a dynamic hypergraph to model user preferences and introduce an intrinsic motivation to train from weak supervision over the director. Finally, to alleviate the bad effect of model bias on the mutual influence between the director and actor, we model the director's option by sampling from a categorical distribution. Extensive experiments demonstrate that DAHCR outperforms state-of-the-art methods.

cs.IR↗

arXiv2023

Multi-view Hypergraph Contrastive Policy Learning for Conversational Recommendation

Conversational recommendation systems (CRS) aim to interactively acquire user preferences and accordingly recommend items to users. Accurately learning the dynamic user preferences is of crucial importance for CRS. Previous works learn the user preferences with pairwise relations from the interactive conversation and item knowledge, while largely ignoring the fact that factors for a relationship in CRS are multiplex. Specifically, the user likes/dislikes the items that satisfy some attributes (Like/Dislike view). Moreover social influence is another important factor that affects user preference towards the item (Social view), while is largely ignored by previous works in CRS. The user preferences from these three views are inherently different but also correlated as a whole. The user preferences from the same views should be more similar than that from different views. The user preferences from Like View should be similar to Social View while different from Dislike View. To this end, we propose a novel model, namely Multi-view Hypergraph Contrastive Policy Learning (MHCPL). Specifically, MHCPL timely chooses useful social information according to the interactive history and builds a dynamic hypergraph with three types of multiplex relations from different views. The multiplex relations in each view are successively connected according to their generation order.

cs.IR↗

arXiv2023

TREA: Tree-Structure Reasoning Schema for Conversational Recommendation

Conversational recommender systems (CRS) aim to timely trace the dynamic interests of users through dialogues and generate relevant responses for item recommendations. Recently, various external knowledge bases (especially knowledge graphs) are incorporated into CRS to enhance the understanding of conversation contexts. However, recent reasoning-based models heavily rely on simplified structures such as linear structures or fixed-hierarchical structures for causality reasoning, hence they cannot fully figure out sophisticated relationships among utterances with external knowledge. To address this, we propose a novel Tree structure Reasoning schEmA named TREA. TREA constructs a multi-hierarchical scalable tree as the reasoning structure to clarify the causal relationships between mentioned entities, and fully utilizes historical conversations to generate more reasonable and suitable responses for recommended results. Extensive experiments on two public CRS datasets have demonstrated the effectiveness of our approach.

cs.AI↗

arXiv2023

Multi-Modal Self-Supervised Learning for Recommendation

The online emergence of multi-modal sharing platforms (eg, TikTok, Youtube) is powering personalized recommender systems to incorporate various modalities (eg, visual, textual and acoustic) into the latent user representations. While existing works on multi-modal recommendation exploit multimedia content features in enhancing item embeddings, their model representation capability is limited by heavy label reliance and weak robustness on sparse user behavior data. Inspired by the recent progress of self-supervised learning in alleviating label scarcity issue, we explore deriving self-supervision signals with effectively learning of modality-aware user preference and cross-modal dependencies. To this end, we propose a new Multi-Modal Self-Supervised Learning (MMSSL) method which tackles two key challenges. Specifically, to characterize the inter-dependency between the user-item collaborative view and item multi-modal semantic view, we design a modality-aware interactive structure learning paradigm via adversarial perturbations for data augmentation. In addition, to capture the effects that user's modality-aware interaction pattern would interweave with each other, a cross-modal contrastive learning approach is introduced to jointly preserve the inter-modal semantic commonality and user preference diversity. Experiments on real-world datasets verify the superiority of our method in offering great potential for multimedia recommendation over various state-of-the-art baselines. The implementation is released at: https://github.com/HKUDS/MMSSL.

cs.IR↗

arXiv2023

On the energy and helicity conservation of the incompressible Euler equations

In this paper, we are concerned with the minimal regularity of weak solutions implying the law of balance for both energy and helicity in the incompressible Euler equations. In the spirit of recent works due to Berselli [5] and Berselli-Georgiadis [6], it is shown that the energy of weak solutions is invariant if $v\in L^{p}(0,T;B^{\frac1p}_{\frac{2p}{p-1},c(\mathbb{N})} )$ with $1<p\leq3$ and the helicity is conserved if $v\in L^{p}(0,T;B^{\frac2p}_{\frac{2p}{p-1},c(\mathbb{N})} )$ with $2<p\leq3 $ for both the periodic domain and the whole space, which generalizes the classical work of Cheskidov-Constantin-Friedlander-Shvydkoy in [10]. This indicates the role of the time integrability, spatial integrability and differential regularity of the velocity in the conserved quantities of weak solutions of the ideal fluid.

math.AP↗

arXiv2023

Multi-View Attention Learning for Residual Disease Prediction of Ovarian Cancer

In the treatment of ovarian cancer, precise residual disease prediction is significant for clinical and surgical decision-making. However, traditional methods are either invasive (e.g., laparoscopy) or time-consuming (e.g., manual analysis). Recently, deep learning methods make many efforts in automatic analysis of medical images. Despite the remarkable progress, most of them underestimated the importance of 3D image information of disease, which might brings a limited performance for residual disease prediction, especially in small-scale datasets. To this end, in this paper, we propose a novel Multi-View Attention Learning (MuVAL) method for residual disease prediction, which focuses on the comprehensive learning of 3D Computed Tomography (CT) images in a multi-view manner. Specifically, we first obtain multi-view of 3D CT images from transverse, coronal and sagittal views. To better represent the image features in a multi-view manner, we further leverage attention mechanism to help find the more relevant slices in each view. Extensive experiments on a dataset of 111 patients show that our method outperforms existing deep-learning methods.

eess.IV↗

arXiv2023

An Evolution Kernel Method for Graph Classification through Heat Diffusion Dynamics

Autonomous individuals establish a structural complex system through pairwise connections and interactions. Notably, the evolution reflects the dynamic nature of each complex system since it recodes a series of temporal changes from the past, the present into the future. Different systems follow distinct evolutionary trajectories, which can serve as distinguishing traits for system classification. However, modeling a complex system's evolution is challenging for the graph model because the graph is typically a snapshot of the static status of a system, and thereby hard to manifest the long-term evolutionary traits of a system entirely. To address this challenge, we suggest utilizing a heat-driven method to generate temporal graph augmentation. This approach incorporates the physics-based heat kernel and DropNode technique to transform each static graph into a sequence of temporal ones. This approach effectively describes the evolutional behaviours of the system, including the retention or disappearance of elements at each time point based on the distributed heat on each node. Additionally, we propose a dynamic time-wrapping distance GDTW to quantitatively measure the distance between pairwise evolutionary systems through optimal matching. The resulting approach, called the Evolution Kernel method, has been successfully applied to classification problems in real-world structural graph datasets. The results yield significant improvements in supervised classification accuracy over a series of baseline methods.

cs.LG↗

arXiv2023

Sizing Grid-Connected Wind Power Generation and Energy Storage with Wake Effect and Endogenous Uncertainty: A Distributionally Robust Method

Wind power, as a green energy resource, is growing rapidly worldwide, along with energy storage systems (ESSs) to mitigate its volatility. Sizing of wind power generation and ESSs has become an important problem to be addressed. Wake effect in a wind farm can cause wind speed deficits and a drop in downstream wind turbine power generation, which however was rarely considered in the sizing problem in power systems. In this paper, a bi-objective distributionally robust optimization (DRO) model is proposed to determine the capacities of wind power generation and ESSs considering the wake effect. An ambiguity set based on Wasserstein metric is established to characterize the wind power and demand uncertainties. In particular, wind power uncertainty is affected by the wind power generation capacity which is determined in the first stage. Thus, the proposed model is a DRO problem with endogenous uncertainty (or decision-dependent uncertainty). To solve the proposed model, a stochastic programming approximation method based on minimum Lipschitz constants is developed to turn the DRO model into a linear program. Then, an iterative algorithm is built, embedded with methods for evaluating the minimum Lipschitz constants. Case studies demonstrate the necessity of considering wake effect and the effectiveness of the proposed method.

eess.SY↗

arXiv2023

Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Framework

Despite recent success on various tasks, deep learning techniques still perform poorly on adversarial examples with small perturbations. While optimization-based methods for adversarial attacks are well-explored in the field of computer vision, it is impractical to directly apply them in natural language processing due to the discrete nature of the text. To address the problem, we propose a unified framework to extend the existing optimization-based adversarial attack methods in the vision domain to craft textual adversarial samples. In this framework, continuously optimized perturbations are added to the embedding layer and amplified in the forward propagation process. Then the final perturbed latent representations are decoded with a masked language model head to obtain potential adversarial samples. In this paper, we instantiate our framework with an attack algorithm named Textual Projected Gradient Descent (T-PGD). We find our algorithm effective even using proxy gradient information. Therefore, we perform the more challenging transfer black-box attack and conduct comprehensive experiments to evaluate our attack algorithm with several models on three benchmark datasets. Experimental results demonstrate that our method achieves overall better performance and produces more fluent and grammatical adversarial samples compared to strong baseline methods. The code and data are available at \url{https://github.com/Phantivia/T-PGD}.

cs.CL↗