1

Explicit Discovery of Nonlinear Symmetries from Dynamic Data

Symmetry is widely applied in problems such as the design of equivariant networks and the discovery of governing equations, but in complex scenarios, it is not known in advance. Most previous symmetry discovery methods are limited to linear …

Low-Dimension-to-High-Dimension Generalization And Its Implications for Length Generalization

Low-Dimension-to-High-Dimension (LDHD) generalization, a subset of Out-of-Distribution (OOD) generalization, involves training on a low-dimensional subspace and testing in a high-dimensional space. Assuming instances are generated from latent …

Projective Equivariant Network via Second-order Fundamental Differential Invariants

Equivariant networks enhance model efficiency and generalization by embedding symmetry priors into their architectures. However, most existing methods, primarily based on group convolutions and steerable convolutions, face significant limitations …

Affine Steerable Equivariant Layer for Canonicalization of Neural Networks

In the field of equivariant networks, achieving affine equivariance, particularly for general group representations, has long been a challenge. In this paper, we propose the steerable EquivarLayer, a generalization of InvarLayer (Li et al., 2024), by …

Relational Learning in Pre-Trained Models: A Theory from Hypergraph Recovery Perspective

Foundation Models (FMs) have demonstrated remarkable insights into the relational dynamics of the world, leading to the crucial question: how do these models acquire an understanding of world hybrid relations? Traditional statistical learning, …

Affine Equivariant Networks Based on Differential Invariants

Convolutional neural networks benefit from translation equivariance, achieving tremendous success. Equivariant networks further extend this property to other transformation groups. However, most existing methods require discretization or sampling of …

Task-Robust Pre-Training for Worst-Case Downstream Adaptation

Pre-training has achieved remarkable success when transferred to downstream tasks. In machine learning, we care about not only the good performance of a model but also its behavior under reasonable shifts of condition. The same philosophy holds when …

Neural ePDOs: Spatially Adaptive Equivariant Partial Differential Operator Based Networks

Endowing deep learning models with symmetry priors can lead to a considerable performance improvement. As an interesting bridge between physics and deep learning, the equivariant partial differential operators (PDOs) have drawn much researchers' …

On the Lower Bound of Minimizing Polyak-Łojasiewicz Functions

Polyak-Łojasiewicz (PL) [Polyak, 1963] condition is a weaker condition than the strong convexity but suffices to ensure a global convergence for the Gradient Descent algorithm. In this paper, we study the lower bound of algorithms using first-order …

Zeroth-order Optimization with Weak Dimension Dependency

Zeroth-order optimization is a fundamental research topic that has been a focus of various learning tasks, such as black-box adversarial attacks, bandits, and reinforcement learning. However, in theory, most complexity results assert a linear …