Representation Learning

SGA: A novel framework boosts Signed Graph Neural Network performance by addressing graph sparsity and unbalanced triangles, achieving up to 26.2% F1-micro improvement.

Divide-and-conquer predictive coding (DCPC) revolutionizes structured Bayesian inference by achieving superior performance in high-dimensional problems while remaining biologically plausible.

Supervised Independent Subspace PCA (sisPCA) disentangles interpretable factors in high-dimensional data by leveraging supervision to maximize subspace dependence on target variables while minimizing …

Boosting structured representation learning, a novel Discrete Dictionary-based Decomposition (D3) layer significantly improves systematic generalization in TPR-based models by efficiently decomposing …

Diffusion models with cross-attention: a powerful inductive bias for effortless disentanglement!

DECRL: A novel deep learning approach for temporal knowledge graph representation learning, capturing high-order correlation evolution and outperforming existing methods.

Researchers developed semantic RSMs, a novel approach to measure semantic similarity in neural networks, improving image retrieval and aligning network representations with predicted class probabiliti…

Node2Vec, a popular network embedding method, is proven to consistently recover community structure in stochastic block models, paving the way for more reliable unsupervised community detection.

Zero-shot learning models often fail in real-world scenarios due to unseen class distribution shifts. This work introduces a novel algorithm that learns robust representations by creating synthetic d…

CtrlNS: A novel framework for causal temporal representation learning tackles the challenge of nonstationary time series by leveraging sparse transition assumptions, achieving improved accuracy in ide…

Pre-trained transformer models can predict human smell perception by encoding odorant chemical structures, aligning with expert labels, continuous ratings, and similarity assessments.

Bisimulation metrics and optimal transport distances are equivalent and can be computed efficiently using a novel Sinkhorn Value Iteration algorithm.

A novel model reveals how hippocampal-entorhinal circuits use compositional coding and modular attractor networks to enable robust and flexible spatial representation, advancing our understanding of c…

High-degree representations significantly boost the expressiveness of E(3)-equivariant GNNs, overcoming limitations of lower-degree models on symmetric structures, as demonstrated theoretically and em…

Latent Mutual Information (LMI) approximation accurately estimates mutual information in high-dimensional data using low-dimensional learned representations, solving a critical problem in various scie…

Deep learning model interpretability improved via Sparse Rate Reduction (SRR), showing improved generalization and offering principled model design.

Hadamard-derived Linear Binding (HLB): A novel, efficient vector symbolic architecture surpassing existing methods in classical AI tasks and deep learning applications.