Abstract:Graph contrastive learning is a general learning paradigm excelling at capturing invariant information from diverse perturbations in graphs. Recent works focus on exploring the structural rationale from graphs, thereby increasing the discriminability of the invariant information. However, such methods may incur in the mis-learning of graph models towards the interpretability of graphs, and thus the learned noisy and task-agnostic information interferes with the prediction of graphs. To this end, with the purpose of exploring the intrinsic rationale of graphs, we accordingly propose to capture the dimensional rationale from graphs, which has not received sufficient attention in the literature. The conducted exploratory experiments attest to the feasibility of the aforementioned roadmap. To elucidate the innate mechanism behind the performance improvement arising from the dimensional rationale, we rethink the dimensional rationale in graph contrastive learning from a causal perspective and further formalize the causality among the variables in the pre-training stage to build the corresponding structural causal model. On the basis of the understanding of the structural causal model, we propose the dimensional rationale-aware graph contrastive learning approach, which introduces a learnable dimensional rationale acquiring network and a redundancy reduction constraint. The learnable dimensional rationale acquiring network is updated by leveraging a bi-level meta-learning technique, and the redundancy reduction constraint disentangles the redundant features through a decorrelation process during learning. Empirically, compared with state-of-the-art methods, our method can yield significant performance boosts on various benchmarks with respect to discriminability and transferability. The code implementation of our method is available at <a class="link-external link-https" href="https://github.com/ByronJi/DRGCL" rel="external noopener nofollow">this https URL</a>.

NCLDR: Nearest-Neighbor Contrastive Learning with Dual Correlation Loss for Dimensionality Reduction

UDRN: Unified Dimensional Reduction Neural Network for Feature Selection and Feature Projection

UDRN: Unified Dimensional Reduction Neural Network for Feature Selection and Feature Projection.

Deep Manifold Transformation for Dimension Reduction

Consistent Representation Learning for High Dimensional Data Analysis

Coupled locality discriminant analysis with globality preserving for dimensionality reduction

Minimizing Nearest Neighbor Classification Error for Nonparametric Dimension Reduction.

LoRA-NCL: Neighborhood-Enriched Contrastive Learning with Low-Rank Dimensionality Reduction for Graph Collaborative Filtering

Nonparametric Locality Reforming Projection for Linear Discriminative Dimensionality Reduction

Multi-layer Cross-domain Non-negative Matrix Factorization for Cross-scene Dimension Reduction on Hyperspectral Images.

Local Feature Discriminant Projection

Rethinking Dimensional Rationale in Graph Contrastive Learning from Causal Perspective

Non-negative Contrastive Learning

Level set learning with pseudo-reversible neural networks for nonlinear dimension reduction in function approximation

Dual Contrastive Learning Network for Graph Clustering

Nonlinear Regression via Deep Negative Correlation Learning

Semi-Supervised Multi-Label Dimensionality Reduction Learning by Instance and Label Correlations

Nonnegative Representation Based Discriminant Projection for Face Recognition.

Adaptive Local Linear Discriminant Analysis

Simultaneously Learning Neighborship and Projection Matrix for Supervised Dimensionality Reduction

Dimension reduction for covariates in network data