A practical method for constructing equivariant multilayer perceptrons for arbitrary matrix groups

M Finzi, M Welling, AG Wilson - International conference on …, 2021 - proceedings.mlr.press
Symmetries and equivariance are fundamental to the generalization of neural networks on
domains such as images, graphs, and point clouds. Existing work has primarily focused on a …

Equivariance with learned canonicalization functions

SO Kaba, AK Mondal, Y Zhang… - International …, 2023 - proceedings.mlr.press
Symmetry-based neural networks often constrain the architecture in order to achieve
invariance or equivariance to a group of transformations. In this paper, we propose an …

Sym-nco: Leveraging symmetricity for neural combinatorial optimization

M Kim, J Park, J Park - Advances in Neural Information …, 2022 - proceedings.neurips.cc
Deep reinforcement learning (DRL)-based combinatorial optimization (CO) methods (ie,
DRL-NCO) have shown significant merit over the conventional CO solvers as DRL-NCO is …

For sale: State-action representation learning for deep reinforcement learning

S Fujimoto, WD Chang, E Smith… - Advances in …, 2024 - proceedings.neurips.cc
In reinforcement learning (RL), representation learning is a proven tool for complex image-
based tasks, but is often overlooked for environments with low-level states, such as physical …

[HTML][HTML] Symmetry-based representations for artificial and biological general intelligence

I Higgins, S Racanière, D Rezende - Frontiers in Computational …, 2022 - frontiersin.org
Biological intelligence is remarkable in its ability to produce complex behaviour in many
diverse situations through data efficient, generalisable and transferable skill acquisition. It is …

-Equivariant Reinforcement Learning

D Wang, R Walters, R Platt - arXiv preprint arXiv:2203.04439, 2022 - arxiv.org
Equivariant neural networks enforce symmetry within the structure of their convolutional
layers, resulting in a substantial improvement in sample efficiency when learning an …

Trustworthy reinforcement learning against intrinsic vulnerabilities: Robustness, safety, and generalizability

M Xu, Z Liu, P Huang, W Ding, Z Cen, B Li… - arXiv preprint arXiv …, 2022 - arxiv.org
A trustworthy reinforcement learning algorithm should be competent in solving challenging
real-world problems, including {robustly} handling uncertainties, satisfying {safety} …

Coordinate Independent Convolutional Networks--Isometry and Gauge Equivariant Convolutions on Riemannian Manifolds

M Weiler, P Forré, E Verlinde, M Welling - arXiv preprint arXiv:2106.06020, 2021 - arxiv.org
Motivated by the vast success of deep convolutional networks, there is a great interest in
generalizing convolutions to non-Euclidean manifolds. A major complication in comparison …

Sample efficient grasp learning using equivariant models

X Zhu, D Wang, O Biza, G Su, R Walters… - arXiv preprint arXiv …, 2022 - arxiv.org
In planar grasp detection, the goal is to learn a function from an image of a scene onto a set
of feasible grasp poses in $\mathrm {SE}(2) $. In this paper, we recognize that the optimal …

Learning layer-wise equivariances automatically using gradients

T van der Ouderaa, A Immer… - Advances in Neural …, 2024 - proceedings.neurips.cc
Convolutions encode equivariance symmetries into neural networks leading to better
generalisation performance. However, symmetries provide fixed hard constraints on the …