Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Laiwan ChanSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works309 from public data
- Review of Causal Discovery Methods Based on Graphical Models1,219
This paper aims to give a introduction to and a brief review of the computational methods for causal discovery that were developed in the past three decades, including constraint-based and score-based methods and those based on functional causal models, supplemented by some illustrations and applications.
- Inferring causation from time series in Earth system sciences1,074
An overview of causal inference frameworks is given, promising applications and methodological challenges are identified, and a causality benchmark platform is initiated to close the gap between method users and developers.
- Deep Domain Generalization via Conditional Invariant Adversarial Networks848
This work proposes an end-to-end conditional invariant deep domain generalization approach by leveraging deep neural networks for domain-invariant representation learning and proves the effectiveness of the proposed method.
- Domain Adaptation under Target and Conditional Shift717
This work considers domain adaptation under three possible scenarios, kernel embedding of conditional as well as marginal distributions, and proposes to estimate the weights or transformations by reweighting or transforming training data to reproduce the covariate distribution on the test domain.
- On causal and anticausal learning699
The problem of function estimation in the case where an underlying causal model can be inferred is considered, and a hypothesis for when semi-supervised learning can help is formulated, and corroborate it with empirical results.
- On learning invariant representations for domain adaptation576
This paper constructs a simple counterexample showing that, contrary to common belief, the above conditions are not sufficient to guarantee successful domain adaptation, and proposes a natural and interpretable generalization upper bound that explicitly takes into account the aforementioned shift.
- Causal discovery and inference: concepts and recent methodological advances411
The constraint-based approach to causal discovery, which relies on the conditional independence relationships in the data, is presented, and the assumptions underlying its validity are discussed.
- SmartBrush: Text and Shape Guided Object Inpainting with Diffusion Model395
This work proposes a new diffusion-based model named SmartBrush for completing a missing region with an object using both text and shape-guidance and shows that this model outperforms all baselines in terms of visual quality, mask controllability, and background preservation.
- Domain Adaptation with Conditional Transferable Components.374
This paper aims to extract conditional transferable components whose conditional distribution P(T(X)∣Y) is invariant after proper location-scale (LS) transformations, and identifies how P(Y) changes between domains simultaneously.
- 346
- Information-geometric approach to inferring causal directions323
This work defines independence via orthogonality in information space so that it can explicitly describe the kind of dependence that occurs between P"Y and P"X"|"Y making the causal hypothesis ''Y causes X'' implausible.
- Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping272
A geometry-consistent generative adversarial network (Gc-GAN) is developed, which enables one-sided unsupervised domain mapping and reduces the space of possible solutions while keep the correct solutions in the search space.
- 229
- Approximate Kernel-Based Conditional Independence Tests for Fast Non-Parametric Causal Discovery228
Two relaxations are devised called the Randomized Conditional Independence Test (RCIT) and the RCoT which both approximate KCIT by utilizing random Fourier features and return accurate p-values much faster than KCIT in the large sample size context.
- Generalized Score Functions for Causal Discovery212
This paper introduces generalized score functions for causal discovery based on the characterization of general (conditional) independence relationships between random variables, without assuming particular model classes.
- Multi-Source Domain Adaptation: A Causal View212
This paper uses causal models to represent the relationship between the features X and class label Y, and considers possible situations where different modules of the causal model change with the domain, to give an intuitive interpretation of the assumptions underlying certain previous methods.
- Causal-learn: Causal Discovery in Python183
This library focuses on bringing a comprehensive collection of causal discovery methods to both practitioners and researchers, and provides easy-to-use APIs for non-specialists, modular building blocks for developers, detailed documentation for learners, and comprehensive methods for all.
- DeepTrader: A Deep Reinforcement Learning Approach for Risk-Return Balanced Portfolio Management with Market Conditions Embedding182
This paper proposes DeepTrader, a deep RL method to optimize the investment policy, which embeds macro market conditions as an indicator to dynamically adjust the proportion between long and short funds, to lower the risk of market fluctuations, with the negative maximum drawdown as the reward function.
- Causal Discovery from Nonstationary/Heterogeneous Data: Skeleton Estimation and Orientation Determination163
This paper proposes an enhanced constraint-based procedure to detect variables whose local mechanisms change and recover the skeleton of the causal structure over observed variables, and presents a way to determine causal orientations by making use of independence changes in the data distribution implied by the underlying causal model.
- Learning Disentangled Semantic Representation for Domain Adaptation160
This work aims to extract the domain invariant semantic information in the latent disentangled semantic representation (DSR) of the data and devise a dual adversarial network to disentangle these two sets of reconstructed latent variables.
- Probabilistic latent variable models for distinguishing between cause and effect143
A novel method for inferring whether X causes Y or vice versa from joint observations of X and Y is proposed, which considers the hypothetical effect variable to be a function of the hypothetical cause variable and an independent noise term.
- 128
- On Estimation of Functional Causal Models110
This article shows that for any acyclic functional causal model, minimizing the mutual information between the hypothetical cause and the noise term is equivalent to maximizing the data likelihood with a flexible model for the distribution of the noiseTerm, and proposes a Bayesian nonparametric approach based on mutual information minimization.
- 105
- Discovering Temporal Causal Relations from Subsampled Data100
This paper proposes an Expectation-Maximization approach and a variational inference approach to recover temporal causal relations from such subsampled data and proves that if the noise terms are non-Gaussian, the underlying model for the highfrequency data is identifiable from subsampling data under mild conditions.
- 92
- Causal Representation Learning from Multiple Distributions: A General Setting84
It is shown that under the sparsity constraint on the recovered graph over the latent variables and suitable sufficient change conditions on the causal influences, interestingly, one can recover the moralized graph of the underlying directed acyclic graph, and the recovered latent variables and their relations are related to the underlying causal model in a specific, nontrivial way.
- Triad Constraints for Learning Causal Structure of Latent Variables83
This paper designs a form of "pseudo-residual" from three variables, and shows that when causal relations are linear and noise terms are non-Gaussian, the causal direction between the latent variables for the three observed variables is identifiable by checking a certain kind of independence relationship.
- Alzheimer’s disease digital biomarkers multidimensional landscape and AI model scoping review79
This review conducts a bibliometric analysis of 431 studies from five online databases and provides a scoping review of 86 artificial intelligence (AI) models, highlighting the progress and challenges of integrating digital biomarkers into clinical practice.
- Learning causality and causality-related learning: some recent progress75
There are still a number of open challenges with regard to deep learning for natural language processing, and deep learning, when combined with other technologies (reinforcement learning, inference, knowledge), may further push the frontier of the field.
- Modeling Dynamic Missingness of Implicit Feedback for Recommendation.72
A latent variable named “user intent” to govern the temporal changes of item missingness, and a hidden Markov model to represent such a process are proposed to capture the dynamic item missingness and incorporate it into matrix factorization (MF) for recommendation.
- Collaborative Filtering With Social Exposure: A Modular Approach to Social Recommendation68
This paper presents a novel recommendation approach (named SERec) to integrate social exposure into collaborative filtering, and proposes two methods to implement SERec, namely social regularization and social boosting, each with different ways to construct social exposures.
- 66
- Adaptive Task Sampling for Meta-learning66
This paper proposes an adaptive task sampling method, which selects difficult tasks according to class-pair potentials and achieves consistent improvements across different feature backbones, meta-learning algorithms and datasets.
- Multi-domain Causal Structure Learning in Linear Systems.66
This work introduces the approach for finding causal direction in a system comprising two variables and proposes efficient methods for identifying causal direction, and generalizes to the case that there is no such invariance across domains.
- Graph Domain Adaptation: A Generative View62
A disentanglement-based unsupervised domain adaptation method for the graph-structured data is proposed, which applies variational graph auto-encoders to recover these latent variables and disentangles them via three supervised learning modules.
- Scalable Causal Discovery with Score Matching62
This paper demonstrates how to discover the whole causal graph from the second derivative of the log-likelihood in observational non-linear additive Gaussian noise models, and develops a practical algorithm that reduces the complexity of the pruning by a factor proportional to the graph size.
- Understanding Masked Autoencoders via Hierarchical Latent Variable Models57
This work formally characterize and justify existing empirical in-sights and provide theoretical guarantees of MAE, and formulate the underlying data-generating process as a hierarchical latent variable model, and shows that under reasonable assumptions, MAE provably identifies a set of latent variables in the hierarchical model.
- 57
- Causal Discovery from Discrete Data using Hidden Compact Representation.50
This paper assumes a two-stage causal process: the first stage maps the cause to a hidden variable of a lower cardinality, and the second stage generates the effect from the hidden representation, which admits a simple yet compact representation.
- Causal Discovery Using Regression-Based Conditional Independence Tests50
This paper proposes a regression-based CI test to relax the test of x ⊥ y|Z to simpler unconditional independence tests, and proves that f and g can be easily estimated by regression, and shows that existing causal learning algorithms can infer much more causal directions by using the proposed method.
- 49
- Causality Discovery with Additive Disturbances: An Information-Theoretical Perspective49
This work considers causally sufficient acyclic causal models in which the relationship among the variables is nonlinear while disturbances have linear effects, and shows that three principles, namely, the causal Markov condition, minimum disturbance entropy, and mutual independence of the disturbances, are equivalent.
- What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective47
It is demonstrated the importance of matching the mission and the means of different fairness inquiries on the data-generating process, the predicted outcome, and the induced impact, respectively to fulfill the intended purpose.
- 47
- Identification of Time-Dependent Causal Model: a gaussian process treatment47
It is shown that under appropriate assumptions, the causal structure is identifiable according to the formulated model, and a principled way for its estimation is proposed by extending Gaussian Process regression, which enables an automatic way to learn how the causal model changes over time.
- Causal Discovery via Reproducing Kernel Hilbert Space Embeddings47
An uncorrelatedness criterion between p(x) and p(y|x) is defined and, based on this uncor relatedness, asymmetry between the cause and the effect is shown in terms of a certain complexity metric that is less than the complexity metric on p (y) andp(x|y).
- Extensions of ICA for Causality Discovery in the Hong Kong Stock Market47
The proposed methods to perform causality discovery even when LiNGAM is violated are applied to discover the causal relations in the Hong Kong stock market, and the last method works very well.
- LTF: A Label Transformation Framework for Correcting Label Shift45
This paper proposes an end-to-end Label Transformation Framework (LTF) for correcting target shift, which implicitly models the shift of P Y and the conditional distribution P X | Y using neural networks.
- Generative-Discriminative Complementary Learning45
A generative-discriminative complementary learning method that estimates the ordinary labels by modeling both the conditional and instance distributions, which improves the accuracy of predicting ordinary labels and is able to generate high-quality instances in spite of weak supervision.
- Unsupervised Sampling Promoting for Stochastic Human Trajectory Prediction44
This work proposes a novel method, called BOsampler, to adaptively mine potential paths with Bayesian optimization in an unsupervised manner, as a sequential design strategy in which new prediction is dependent on the previously drawn samples.
- Fair Representation Learning for Recommendation: A Mutual Information Perspective43
This paper re-define recommendation fairness with a novel two-fold mutual information objective, and proposes novel mutual information (upper/lower) bounds with contrastive information estimation for model optimization.
- Evolving Semantic Prototype Improves Generative Zero-Shot Learning42
A dynamic semantic prototype evolving (DSP) method is proposed to align the empirically predefined semantic prototypes and the real prototypes for class-related feature synthesis and indicates that evolving semantic prototype explores a virgin field in ZSL.
- 41
- Label-Noise Robust Domain Adaptation.41
This paper is the first to comprehensively investigate how label noise could adversely affect existing domain adaptation methods in various scenarios and theoretically prove that there exists a method that can essentially reduce the side-effect of noisy source labels in domain adaptation.
- 41
- 40
- 39
- 39
- Minimal Nonlinear Distortion Principle for Nonlinear Independent Component Analysis39
This paper proposes the “minimal nonlinear distortion” (MND) principle for tackling the ill-posedness of nonlinear ICA problems, and uses non linear ICA with MND to separate daily returns of a set of stocks in Hong Kong, and the linear causal relations among them are successfully discovered.
- A Versatile Causal Discovery Framework to Allow Causally-Related Hidden Variables38
A novel, versatile framework for causal discovery that accommodates the presence of causally-related hidden variables almost everywhere in the causal network, based on rank information of covariance matrix over observed variables is introduced.
- Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows38
A novel kernel embedding of nonstationary conditional distributions that does not rely on sliding windows is developed and leads to a measure of dependence between the changes of causal modules that can be used to determine the directions of many causal arrows.
- Causal Balancing for Domain Generalization34
This work proposes a balanced mini-batch sampling strategy to transform a biased data distribution into a spurious-free balanced distribution, based on the invariance of the underlying causal mechanisms for the data generation process, and argues that the Bayes optimal classifiers trained on such balanced distribution are minimax optimal across a diverse enough environment space.
- Quantum multi-programming for Grover’s search33
It is proved that this new algorithm increases the rotation angle of the Grover operator which, as a result, increases the success probability and outperforms other variations of Grover's algorithms as well as the canonical Grover’s algorithm.
- Maximum Spatial Perturbation Consistency for Unpaired Image-to-Image Translation33
A universal regularization technique called maximum spatial perturbation consistency (MSPC), which enforces a spatial perturgation function and the translation operator to be commutative and outperforms the state-of-the-art methods on most I2I benchmarks.
- 31
- Specific and Shared Causal Relation Modeling and Mechanism-Based Clustering31
This paper proposes a specific and shared causal model (SSCM), which takes into account the variabilities of causal relations across individuals/groups and leverages their commonalities to achieve statistically reliable estimation.
- Joint extraction of entities and overlapping relations by improved graph convolutional networks29
This paper establishes a novel model to extract entities and relations simultaneously by using improved Multi-task Graph Convolutional networks, called MGCN, and proposes an efficient strategy to map multiple relational labels of a sentence into a unique code.
- Improving Causal Discovery By Optimal Bayesian Network Learning28
It is demonstrated that optimal score-based exhaustive search is remarkably useful for causal discovery: it requires weaker conditions to guarantee asymptotic correctness, and outperforms well-known methods including PC, GES, GSP, and NOTEARS.
- Causal Discovery with Cascade Nonlinear Additive Noise Model28
Simulation results illustrate the power of the proposed method in identifying indirect causal relations across various settings, and experimental results suggest that the proposed model and method greatly extend the applicability of causal discovery based on functional causal models in nonlinear cases.
- Causal Discovery with Linear Non-Gaussian Models under Measurement Error: Structural Identifiability Results.28
This study assumes that the underlining process or the measurement-error free variables follows a linear, non-Guassian causal model, and shows that the so-called ordered group decomposition of the causal model is identifiable, which contains major causal information.
- CaRiNG: Learning Temporal Causal Representation under Non-Invertible Generation Process27
This work establishes an identifiability theory that allows for the recovery of independent latent components even when they come from a nonlinear and non-invertible mix and proposes a principled approach, CaRiNG, to learn the CAusal RepresentatIon of Non-invertible Generative temporal data with identifiability guarantees.
- Identifiable Latent Polynomial Causal Models Through the Lens of Change27
The scope of latent causal models is extended to involve nonlinear causal relationships, represented by polynomial models, and general noise distributions conforming to the exponential family, and a novel empirical estimation method is proposed, grounded in the theoretical finding, that enables learning consistent latent causal representations.
- Offline Reinforcement Learning with Causal Structured World Models27
A practical algorithm, oFfline mOdel-based reinforcement learning with CaUsal Structure (FOCUS), to illustrate the feasibility of learning and leveraging causal structure in offline RL and results show that FOCUS reconstructs the underlying causal structure accurately and robustly.
- 27
- Cloud Atlas: Efficient Fault Localization for Cloud Systems using Language Models and Causal Insight26
Atlas is presented, a novel approach to automatically synthesizing causal graphs for cloud systems that leverages large language models to generate causal graphs using system documentation, telemetry, and deployment feedback, and is complementary to data-driven causal discovery techniques.
- 26
- 25
- Identification of Linear Latent Variable Model with Arbitrary Distribution25
The identifiability result relaxes the non-Gaussianity requirements to only a (hopefully small) subset of variables, and accordingly elegantly extends the application scope of the structural model.
- Counting and Sampling from Markov Equivalent DAGs Using Clique Trees25
This paper proposes a new technique for counting the number of DAGs in a Markov equivalence class based on the clique tree representation of chordal graphs and shows that in the case of bounded degree graphs, the proposed algorithm is polynomial time.
- Semi-supervised Learning in Causal and Anticausal Settings25
The hypothesis that semi-supervised learning can help in an anti-causal setting, but not in a causal setting is formulated, and empirical results corroborate it with empirical results.
- Transfer Learning-Based Outdoor Position Recovery With Cellular Data24
This framework is the first framework that demonstrates the efficacy of applying transfer learning in the Telco outdoor position recovery and introduces three dedicated components: a new coordinate space to divide an area of interest into smaller domains, a similarity measurement to select best source domains, and an adaptation of an existing transfer learning approach.
- Data-Driven Approach to Multiple-Source Domain Adaptation.24
A data-driven method to identify the changing parameters by learning low-dimensional representations of the changing class-conditional distributions across multiple source domains and perform unsupervised domain adaptation is developed.
- Multi-scale Masked 3-D U-Net for Brain Tumor Segmentation24
This paper proposes a multi-scale masked 3D U-Net which captures multi- Scale information by stacking multi- scale images as inputs and incorporating a 3-D Atrous Spatial Pyramid Pooling (ASPP) layer to boost the performance of brain tumor segmentation.
- On the identifiability and estimation of functional causal models in the presence of outcome-dependent selection24
It is shown that in the framework of post-nonlinear causal models, once outcome-dependent selection is properly modeled, the causal direction between two variables is generically identifiable; and some mild conditions under which an additive noise causal model with outcome- dependent selection is to a large extent identifiable.
- Project-Driven Learning-by-Doing Method for Teaching Software Engineering using Virtualization Technology24
The paper shows that learning-by-doing method correlates with the characteristics of different projects, which has resulted in a successful experience as reported by students in an end of a semester survey.
- Residual Similarity Based Conditional Independence Test and Its Application in Causal Discovery23
This paper investigates the independence between two linear combinations under linear non-Gaussian structural equation model and designs a new method called SCIT for CI test, where permutation test is performed to control Type I error rate.
- Region-aware network: Model human’s Top-Down visual perception mechanism for crowd counting23
A novel feedback network with Region-Aware block called RANet is proposed by modeling human's Top-Down visual perception mechanism by introducing a feedback architecture to generate priority maps that provide prior about candidate crowd regions in input images.
- Extended Gaussianization Method for Blind Separation of Post-Nonlinear Mixtures23
The gaussianization method for BSS of PNL mixtures is reviewed and extended to apply to PNL mixture in which the linear mixture is close to gaussian, and the proposed method approximates thelinear mixture using the Cornish-Fisher expansion.
- Testing Independence Between Linear Combinations for Causal Discovery22
The proposed method provides a simpler but more effective way to measure CIs, with only calculating the 1-st to 4-th moments of the input variables, and outperforms kernel-based methods in terms of both speed and accuracy.
- Replication Package for Article: AKG: Automatic Kernel Generation for Neural Processing Units using Polyhedral Transformations22
AKG is presented, a tensor compiler for NPUs that leverages polyhedral schedulers to perform a much wider class of transformations, and extends the semantics of the polyhedral representation to combine complex tiling techniques and hierarchical fusion strategies.
- ICA with Sparse Connections22
This paper proposes natural gradient algorithms for ICA with a sparse separation matrix, as well as a sparse mixing matrix, where the sparsity of the matrix is achieved by applying certain penalty functions to its entries.
- Improving the Expressiveness of K-hop Message-Passing GNNs by Injecting Contextualized Substructure Information21
The limitation of K-hop message-passing GNNs is discussed and a substructure encoding function to uplift the expressive power of any K- Hop message- passing GNN is proposed, which is provably more powerful than previous works on K-Hop graph neural networks and 1-WL subgraph GNNS.
- 21
- Emerging Synergies in Causality and Deep Generative Models: A Survey20
A comprehensive review of techniques that incorporate causal principles within DGMs, methods for identifying causal relationships through generative modeling, and emerging research frontier of causality in LLMs is provided.
- Partial Identifiability for Domain Adaptation20
This work proposes a practical domain adaptation framework called iMSDA, which outperforms state-of-the-art domain adaptation algorithms on benchmark datasets, demonstrating the effectiveness of this property-based framework.
- Feature Expansion for Graph Neural Networks20
This paper decomposes graph neural networks into determined feature spaces and trainable weights, providing the convenience of studying the feature space explicitly using matrix space analysis, and theoretically finds that thefeature space tends to be linearly correlated due to repeated aggregations.
- Relevance attack on detectors20
This paper generates the first adversarial dataset for object detection and instance segmentation, i.e., Adversarial Objects in COntext (AOCO), which helps to quickly evaluate and improve the robustness of detectors.
- On the Identification of Temporally Causal Representation with Instantaneous Dependence19
This work establishes identifiability results of the latent causal process based on sufficient variability and the sparse influence constraint by employing contextual information of time series data and incorporates a temporally variational inference architecture to estimate the latent variables and a gradient-based sparsity regularization to identify the latent causal process.
- Generalized Independent Noise Condition for Estimating Causal Structure with Latent Variables19
The proposed Generalized Independent Noise (GIN) condition for linear non-Gaussian acyclic causal models that incorporate latent variables is proposed, which establishes the independence between a linear combination of certain measured variables and some other measured variables.
- 18
- Kernel-Based Nonlinear Independent Component Analysis18
The kernel-based nonlinear independent component analysis method, which consists of two separate steps, utilizes the minimal nonlinear distortion (MND) principle for regularization, in addition to the smoothness regularizer, to overcome the ill-posedness in nonlinear ICA solutions.
- 17
- 17
- A general linear non-Gaussian state-space model: Identifiability, identification, and applications17
It is proved that this model is fully identiable and an ecient two-step method for parameter estimation is proposed: one extracts the subspace of the latent processes based on the temporal information of the data, and then performs multichannel blind deconvolution, making use of both the temporal Information and non-Gaussianity.
- Testability of Instrumental Variables in Linear Non-Gaussian Acyclic Causal Models16
A necessary condition for detecting variables that cannot serve as instrumental variables that is designed with a single instrumental variable and a method to select the set of candidate instrumental variables given observational data is developed.
- ACAMDA: Improving Data Efficiency in Reinforcement Learning through Guided Counterfactual Data Augmentation15
ACAMDA (Adversarial Causal Modeling for Data Augmentation), a novel framework that integrates two causality-based tasks: causal structure recovery and counterfactual estimation, outperforms existing methods, particularly when applied to novel and unseen domains.
- 15
- ICA by PCA Approach: Relating Higher-Order Statistics to Second-Order Moments15
An elegant way to incorporate higher-order statistics implicitly in the form of second-order moments is given, and it is shown that ICA can be performed by PCA following a simple transformation.
- 14
- Invariant Action Effect Model for Reinforcement Learning14
This paper shows that a better representation can be learned and potentially improves the sample efficiency and the generalization ability of the learned policy and proposes Invariant Action Effect Model (IAEM), which captures the invariance in action effects, where the effect of an action is represented as the residual of representations from neighboring states.
- 14
- Four changes for efficiency and practicality on previous 3D MT NLCG inversion algorithm14
A parallel inversion algorithm for 3D magnetotelluric sounding based on the previous preconditioned nonlinear conjugate gradient method with an improved preconditionsing scheme is presented, and the forward iteration limit is changed.
- Convolutive blind source separation by efficient blind deconvolution and minimal filter distortion14
This paper presents a computationally efficient approach to multichannel blind deconvolution (MBD) and shows that with this principle, each recovered source is approximately the principal component of the contributions of this source to all observations.
- MuGSI: Distilling GNNs with Multi-Granularity Structural Information for Graph Classification13
MuGSI is introduced, a novel KD framework that employs Multi-granularity Structural Information for graph classification and proposes to incorporate a node feature augmentation component, thereby enhancing the expressiveness of the student MLPs and making them more capable learners.
- Design and application of coal gangue sorting system based on deep learning13
A deep learning-based, non-contact gangue recognition and pneumatic intelligent sorting system that achieves a gangue identification accuracy exceeding 97%, a sorting rate above 91%, and a separation time of less than 3 s from identification to separation, thereby effectively enhancing raw coal purity.
- 13
- HCVP: Leveraging Hierarchical Contrastive Visual Prompt for Domain Generalization12
This work introduces the novel Hierarchical Contrastive Visual Prompt (HCVP) methodology, which represents a significant advancement in the field, setting itself apart with a unique generative approach to prompts, alongside an explicit model structure and specialized loss functions.
- 12
- 12
- 12
- 12
- 11
- Deep State-Space Generative Model For Correlated Time-to-Event Predictions11
A deep latent state-space generative model is proposed to capture the interactions among different types of correlated clinical events by explicitly modeling the temporal dynamics of patients' latent states by developing a new general discrete-time formulation of the hazard rate function.
- 11
- 10
- GGATB-LSTM: Grouping and Global Attention-based Time-aware Bidirectional LSTM Medical Treatment Behavior Prediction10
A Grouping and Global Attention based Time-aware Bidirectional Long Short-Term Memory (GGATB-LSTM) model is proposed to achieve medical treatment behavior prediction and is evaluated by using real dataset to conclude that GGATBs is better than other methods.
- Generalized Independent Noise Condition for Estimating Linear Non-Gaussian Latent Variable Graphs.10
This paper considers Linear, Non-Gaussian Latent variable Models (LiNGLaMs), in which latent confounders are also causally related, and proposes a Generalized Independent Noise (GIN) condition and shows that GIN helps locate latent variables and identify their causal structure, including causal directions.
- 10
- 10
- Single image deblurring using novel image prior constraints10
A probabilistic model is established by enforcing inspired image prior constraints and an advanced iterative scheme that alternates between blur kernel estimation and non-blind image restoration and an alternation half-quadratic algorithm to solve the non-convex cost function is presented.
- OCRT: Boosting Foundation Models in the Open World with Object-Concept-Relation Triad9
A novel framework is proposed, the Object-Concept-Relation Triad (OCRT), that enables FMs to extract sparse, high-level concepts and intricate relational structures from raw visual inputs and can substantially boost the generalizability and robustness of SAM and CLIP across multiple downstream tasks.
- Long-Term Fairness Inquiries and Pursuits in Machine Learning: A Survey of Notions, Methods, and Challenges9
This survey reviews existing literature on long-term fairness from different perspectives and presents a taxonomy for long-term fairness studies, highlighting key challenges and consider future research directions, analyzing both current issues and potential further explorations.
- Local Causal Discovery with Linear non-Gaussian Cyclic Models9
This work presents a general, unified local causal discovery method with linear non-Gaussian models, whether they are cyclic or acyclic, and extends the application of independent component analysis from the global context to independent subspace analysis, enabling the exact identification of the equivalent local directed structures and causal strengths from the Markov blanket of the target variable.
- Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factors9
Tier Balancing is proposed, a technically more challenging but more natural notion to achieve in the context of long-term, dynamic fairness analysis, that goes one step further, capturing behind-the-scenes situation changes on the unobserved latent causal factors that directly carry out the influence from the current decision to the future data distribution.
- 9
- 9
- CausalNLP Tutorial: An Introduction to Causality for Natural Language Processing9
This tutorial introduces the fundamentals of causal discovery and causal effect estimation to the natural language processing (NLP) audience, provides an overview of causal perspectives to NLP problems, and aims to inspire novel approaches to N LP further.
- PRNet9
A deep neural network (DNN)-based position recovery framework, namely PRNet, which can ensemble the power of CNN, sequence model LSTM, and two attention mechanisms to learn local, short- and long-term spatio-temporal dependencies from input MR samples is proposed.
- Nonlinear independent component analysis with minimal nonlinear distortion9
This paper proposes to solve nonlinear ICA with the "minimal nonlinear distortion" principle by incorporating a regularization term to minimize the mean square error between the mixing mapping and the best-fitting linear one.
- Reflection-Window Decoding: Text Generation with Selective Refinement8
This paper proposes an approach that incorporates a sliding reflection window and a pausing criterion, such that refinement and generation can be carried out interchangeably as the decoding proceeds, and its extensive experimental results demonstrate the effectiveness of the approach.
- 8
- 8
- Causal Discovery with Confounding Cascade Nonlinear Additive Noise Models8
Simulation results illustrate the power of the proposed method in identifying indirect causal relations across various settings, and experimental results suggest that the proposed model and method greatly extend the applicability of causal discovery based on functional causal models in nonlinear cases.
- Outcome-based school-enterprise cooperative software engineering training8
This paper proposes an outcome-based school-enterprise cooperative software engineering training to increase student engagement of engineering practice, and address some real complicated engineering issue.
- Unmixing for Causal Inference: Thoughts on McCaffrey and Danks8
An algorithm is given that addresses the problem of discovering causal relations in data drawn from a mixture of distributions commonly assumed in fMRI studies and finds that in testing, it can accurately separate data from mixed distributions.
- 8
- 8
- Functional linear non-Gaussian acyclic model for causal discovery7
It is demonstrated why the original LiNGAM fails to handle these inherently infinite-dimensional datasets, and theoretical guarantees of the identifiability of the causal relationship among non-Gaussian random vectors and even random functions in infinite-dimensional Hilbert spaces are established.
- Generalized Precision Matrix for Scalable Estimation of Nonparametric Markov Networks7
This work characterize the conditional independence structure in general distributions for all data types with a Generalized Precision Matrix and unify all cases under the same umbrella of a regularized score matching framework, thus giving rise to a Markov network structure learning algorithm in one of the most general settings.
- Efficient Object Detection in SAR Images Based on Computation-Aware Neural Architecture Search7
This paper proposes Darts_Tiny, a novel differentiable neural architecture search model, to design dedicated convolutional neural networks automatically, and demonstrates that the neural networks win by at least an order of magnitude in terms of model complexity compared with SOTA lightweight models.
- 7
- 7
- 7
- Enhancement of Source Independence for Blind Source Separation7
This paper investigates ways to find the optimal filter for enhancing source independence in two scenarios, if none of the sources is known and if some source signals are available, and adaptively estimate the filter and the de-mixing matrix simultaneously by minimizing the mutual information between outputs.
- Synergy Between Sufficient Changes and Sparse Mixing Procedure for Disentangled Representation Learning6
An identifiability theory with less restrictive constraints regarding distribution changes and the sparse mixing procedure is proposed, enhancing applicability to real-world scenarios and developing an estimation framework incorporating a domain encoding network and a sparse mixing constraint.
- 6
- Salesforce CausalAI Library: A Fast and Scalable Framework for Causal Analysis of Time Series and Tabular Data6
The Salesforce CausalAI Library is introduced, an open-source library for causal analysis using observational data that supports causal discovery and causal inference for tabular and time series data, of discrete, continuous and heterogeneous types.
- 6
- 6
- Adversarial orthogonal regression: Two non-linear regressions for causal inference6
The proposed Adversarial Orthogonal Regression methods can be used as subroutines to address several learning problems in causality, such as causal direction determination and causal structure learning and causal model estimation.
- 6
- Introduction to the Special Section on Advances in Causal Discovery and Inference6
This special issue collects seven articles that study the fundamental problems in causal discovery and inference and present novel solutions for false discovery control in structure learning, causal relationship detection in simulation models, causal structure search in the presence of latent confounders, the shortest causal path discovery by local search, and conditional independence test for causal structure learning.
- 6
- Causal discovery with scale-mixture model for spatiotemporal variance dependencies6
This paper proposes a probabilistic causal scale-mixture model with spatiotemporal variance dependencies to represent a specific type of generating mechanism of the observations and proves the identifiability of this model under the non-Gaussian assumption on the innovation processes.
- Agent interface enhancement6
This work presents multiagent interface enhancement that relieves agent designers from the burden of model efficiency so that multiagent graphical models become more accessible.
- 6
- Permutation-Based Rank Test in the Presence of Discretization and Application in Causal Discovery with Mixed Data5
MPRT can effectively control the Type I error in the presence of discretization while previous methods cannot, and is validated by extensive experiments on synthetic data and real-world data to demonstrate its effectiveness as well as applicability in causal discovery.
- Classifying Treatment Responders: Bounds and Algorithms5
Extensive experiments on semi-synthetic and real-world datasets show that the proposed CTRL algorithm better predicts treatment responders and adaptively trades off false-positives and false-negatives with varying weight coefficients.
- Explainable Recommender With Geometric Information Bottleneck5
The proposed model significantly improves the interpretability of a variational recommender using the Wasserstein distance while achieving performance comparable to existing content-based recommender systems in terms of recommendation behaviours.
- Differentiable Causal Discovery For Latent Hierarchical Causal Models5
This work develops a novel differentiable causal discovery algorithm that efficiently estimates the structure of nonlinear latent hierarchical causal models and demonstrates its practical utility by learning interpretable hierarchical latent structures from high-dimensional image data and on downstream tasks.
- 5
- Nonstationary Time Series Forecasting via Unknown Distribution Adaptation5
The IDEA model, which incorporates an autoregressive hidden Markov model to estimate latent environments and modular prior networks to identify latent states, out-performs several latest nonstationary forecasting methods on various benchmark datasets, high-lighting its advantages in real-world scenarios.
- 5
- Image reconstruction of electrostatic tomography based on the improved residual network5
An improved ResNet-34 network (P-ResNet), which consists of an input layer, a residual feature extraction layer, and an output layer, is presented, which achieves high accuracy in EST image reconstruction tasks.
- 5
- 5
- Correlations among positive psychological capital, research motivation, and research ability by master's degree nursing students: A structural equation modeling5
The positive psychological capital and research motivation of master's degree nursing students are positively correlated with research ability and research motivation significantly mediates the relationship between positive psychological capital and research ability.
- 5
- Identifying Weight-Variant Latent Causal Models5
This work analysis three intrinsic properties in latent space, including transitivity, permutation and scaling, and proposes a novel method, termed Structural caUsAl Variational autoEncoder (SuaVE), which directly learns latent causal variables, together with the mapping from the latent causal variable to the observed ones.
- 5
- 5
- 5
- 5
- The Evaluation of Discovery: Models, Simulation and Search through “Big Data”5
A central theme in western philosophy was to find formal methods that can reliably discover empirical relationships and their explanations from data assembled from experience, but to show that the model/simulation strategy is not confined to causal inference, two classification problems from astrophysics are considered: identifying exoplanets and identifying dark matter concentrations.
- 5
- 5
- 5
- Nonparametric Factor Analysis and Beyond4
This work shows that, under suitable conditions, the generative model is identifiable up to certain submanifold indeterminacies even in the presence of non-negligible noise, and proves that latent variables of the general nonlinear models are identifiable up to trivial indeterminacies.
- Horizontal and Vertical Federated Causal Structure Learning via Higher-order Cumulants4
This work comprehensively considers causal structure learning methods under both horizontal and vertical federated settings via higher-order cumulants and provides the identification theories and methods for learning causal structure in the horizontal and vertical federal setting via higher-order cumulants.
- 4
- 4
- A Conditional Independence Test in the Presence of Discretization4
A conditional independence test specifically designed to accommodate the presence of a discretization of latent variables, and design the bridge equations to recover the parameter reflecting the statistical information of the underlying latent continuous variables.
- 4
- 4
- Conditional Independence Test Based on Residual Similarity4
A new way to test CI based on the similarity between residuals, called SCIT — the abbreviation of Similarity-based CI Testing is provided, and two versions of the proposal are developed, called Kernel-SCIT and Neural- SCIT, respectively.
- 4
- Outdoor Position Recovery From Heterogeneous Telco Cellular Data4
A feature extraction module is developed to precisely learn local-, short- and long-term spatio-temporal locality from heterogeneous MR samples from multi-task learning-based deep neural network (DNN) framework.
- 4
- 4
- 4
- Causal Discovery and Hidden Driving Force Estimation from Nonstationary/Heterogeneous Data4
A principled framework for causal discovery from NOnstationary/heterogeneous Data (CD-NOD) is developed, which addresses two important questions: an enhanced constraint-based procedure to detect variables whose local mechanisms change and recover the skeleton of the causal structure over observed variables and the tight connection between nonstationarity/heterogeneity and soft intervention in causal discovery.
- Preference-driven yield-and-quality optimization for high-sulfur gas sweetening process by extreme learning machine model4
Data mining methods are used to explore the relationship between parameters and output variables of high-sulfur gas sweetening process and apply it to optimize the yield and quality and show that the established ELM model is in good agreement with the actual operation data.
- Algorithm for camera parameter adjustment in multicamera systems4
A new algorithm is proposed to improve the accuracy and consistency of Multicamera systems by adjusting the camera parameters by assuming that the distribution of the measured point positions follows the Gaussian mixture model.
- Oblique decision tree induction using multimembered evolution strategies4
A new construction algorithm for binary oblique decision tree classifier, MESODT, which combines Multimembered evolution strategies integrated with the perceptron algorithm as the optimization algorithm to find the appropriate split that minimizes the evaluation function at each node of a decision tree.
- Higher-Order Causal Structure Learning with Additive Models3
This work focuses on extending the causal additive model (CAM) to additive models with higher-order interactions, and develops an extension of the greedy CAM algorithm which can handle the more complex hyper DAG search space and demonstrate its empirical usefulness in synthetic experiments.
- PersonaX: Multimodal Datasets with LLM-Inferred Behavior Traits3
PersonaX, a curated collection of multimodal datasets designed to enable comprehensive analysis of public traits across modalities, and introduces a novel causal representation learning (CRL) framework tailored to multimodal and multi-measurement data, providing theoretical identifiability guarantees.
- 3
- 3
- Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Models3
This work proposes the first score-based greedy search method for the identification of structure involving latent variables with identifiability guarantees and designs Latent variable Greedy Equivalence Search, a greedy search algorithm for this class of model with well-defined operators, which search very efficiently over the graph space to find the optimal structure.
- Procedural Fairness Through Decoupling Objectionable Data Generating Components3
A framework is proposed that decouples the objectionable data generating components from the neutral ones by utilizing reference points and the associated value instantiation rule and highlights the necessity of preventing disguised procedural unfairness.
- Latent Covariate Shift: Unlocking Partial Identifiability for Multi-Source Domain Adaptation3
It is shown that although (completely) identifying the proposed latent causal model is challenging, the latent content variable can be identified up to scaling by using its dependence with labels from source domains, together with the identifiability conditions of nonlinear ICA.
- 3
- 3
- 3
- 3
- 3
- Reviewer Assignment Strategy of Peer Assessment: Towards Managing Collusion in Self-assignment3
This research aims to propose a model of collusion management that applies cooccurrence network as a means to identify collusion and is implemented as a software module in a peer code review system called EduPCR.
- 3
- Exploring model-driven engineering method for teaching software engineering3
The organisation of both projects are served as an inspiration for higher schools that want to develop a software engineering course and some lessons learned from the experiences are concluded.
- On Causal Discovery in the Presence of Changing Causal Models3
An enhanced constraint-based method to detect variables whose local mechanisms are nonstationary and recover the skeleton of the causal structure over observed variables is proposed and a method for visualizing the nonstationarity of causal modules is developed.
- 3
- 2
- A Sample Efficient Conditional Independence Test in the Presence of Discretization2
This paper finds that the independence relationships of latent continuous variables can be established by addressing an over-identifying restriction problem with Generalized Method of Moments (GMM), and derives an appropriate test statistic and establishes its asymptotic distribution correctly reflecting CI by leveraging nodewise regression.
- Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens2
This paper revisits reasoning tasks from a causal perspective, seeking to understand their behavior in latent space and to offer insights for addressing their challenges, and introduces a framework that incorporates the estimated latent variables as feedback into the selection mechanism, thereby facilitating the learning of dense dependencies among latent representations.
- 2
- A Covariance Adaptive Student’s t-Based Kalman Filter2
The Gaussian mixture model (GMM) is used to construct an adaptive measurement covariance matrix as a way to adaptively balance the predicted and measurement data.
- Advancing Counterfactual Inference through Nonlinear Quantile Regression2
This paper re-cast counterfactual reasoning as an extended quantile regression problem using neural networks and makes it possible to develop the generalization ability of the estimated counterfactual outcome to unseen data and provide an upper bound on the generalization error.
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Independent Factor Reinforcement Learning for Portfolio Management2
Experimental results and analysis are given to show that the proposed method has better performance when compare to two alternative portfolio management systems.
- 2
- 1
- 1
- 1
- Learning by Analogy: A Causal Framework for Composition Generalization1
This work introduces a hierarchical data-generating process that naturally encodes different levels of concepts and their interaction mechanisms and proves that this latent hierarchical structure is provably recoverable from observable data like text-image pairs, a necessary step for learning such a generative process.
- Continual Unsupervised Generative Modelling via Online Optimal Transport1
This paper proposes the Sinkhorn-based Dual Dynamic Memory method, by considering Sinkhorn distance as an optimal transport measure, for evaluating the significance of the data to be stored in the memory buffer and proposes a novel dynamic model expansion mechanism to automatically increase the model's capacity whenever necessary.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- A Software/Hardware Co-design Local Irregular Sparsity Method for Accelerating CNNs on FPGA1
A software/hardware co-design approach to bring local irregular sparsity into CNNs and design a row-wise computing engine, RConv Engine, to achieve workload balance and remarkable speedup.
- Dense subgraph mining kmax-truss optimization algorithm1
A dynamic adaptive step size method is proposed that uses the clique to analyze the intermediate results, ending the iterative process early and improving the execution efficiency, and results show that 30% iterations can be reduced.
- Computational causal discovery: Advantages and assumptions1
This paper will focus on some of the emerging computational methods for finding causal relations from non-experimental data and attempt to complement Woodward's contribution with discussions on how these methods are connected to the interventionist theory of causality.
- 1
- 1
- 1
- 1
- 1
- Unpaired data empowers association tests1
This work presents a new method called the Semi-paired Association Test (SAT) that makes use of both paired and unpaired data, and identifies an association between the high-dimensional characterization of Computed Tomography chest images and blood biomarkers as well as the expression of dozens of genes involved in the immune system.
- Sparse Discriminant Preserving Projections for Face Recognition1
Since SDPP only takes into account the within-class sparsity reconstructive relationship and the between-class scatter information, it not only well preserves the sparse reconstruction relationship, but also improves classification performance.
- 1
- Power load prediction method based on kernel extreme learning machine with t-distribution variation bat algorithm1
AMBA-KLEM model is used to predict load of Ningxia Power Grid, and the results show that the model has higher load prediction accuracy.
- 1
- Learning Vector Autoregressive Models With Latent Processes1
It is shown that the support of transition matrix among the observed processes and lengths of all latent paths between any two observed processes can be identified successfully under some conditions on the VAR model.
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- MixAR: Mixture Autoregressive Image Generation–
MixAR is a novel framework that leverages mixture training paradigms to inject discrete tokens as prior guidance for continuous autoregressive prediction and proposes Training-Inference Mixture (TI-Mix) to achieve consistent training and generation distributions.
- –
- –
- –
- Multi-attribute fusion generation via text-to-image customization–
The cross-attention mechanism is used to effectively capture the hierarchical relationships between text and images, guide image-generation through diffusion model in the process of multi-attribute fusion and present the effect of customized images with multi-attribute fusion.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- ICTNET at Trec 2019 Incident Streams Track.–
This paper deals with this learning task with neural network in the view of classification problem to cope with crisis management via social medial.
- Mining High-Utility Itemsets Based on Multiple Minimum Support and Multiple Minimum Utility Thresholds–
This paper addresses the issue of how to discover high utility itemsets from a transactional database by combining two different kinds of thresholds used by HUIM-MMU and MHU-Growth by using two algorithms namely H UIM-MMSU and HU IM-IMMSU.
- –
- –
- –
- –
- –
- –
- Point Cloud Data Organization Algorithm Based on Clustering–
A new algorithm for point cloud organization based on improved Kmeans is provided, which can be store as a tree and verified by the experiment results.
- A Multiple Moving Object Segmentation Algorithm Based on Background Modeling and Adaptive Clustering–
A multiple moving object segmentation algorithm based on Background Modeling and Adaptive Clustering (named as BMAC) and the use of space connectivity information improve the accuracy of clustering segmentation, comparison and analysis show that the proposed algorithm is feasible, rapid and effective.
- An identifiability result for SVAR processes with hidden components assuming non-Gaussian noise–
It is shown that under certain assumptions, using only X, one can fully identify that part of A that captures the interaction between the components of X, which may help to improve causal analysis of time series in certain cases.
- An Evaluation Method for Water Quality with Similarity Measures of Vague Sets–
The new formulas for similarity measures between Vague sets will be used as pattern recognition method based on similarity measures in water quality evaluation and this application example shows that this method is not only convenient in calculation, but also practical in results.
- Robust Facial Feature Tracking Using an Autoregressive Model and Optical Flow–
Chen, J., Zhang, K., Gong, Y., Tiddeman, B. (2012) Robust Facial Feature Tracking Using an Autoregressive Model and Optical Flow.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.