Equivariant Neural Tangent Kernels

• URL: http://arxiv.org/abs/2406.06504v2
• Date: Fri, 31 Jan 2025 14:56:51 GMT
• Title: Equivariant Neural Tangent Kernels
• Authors: Philipp Misof, Pan Kessel, Jan E. Gerken,
• Abstract summary: neural tangent kernels (NTKs) have emerged as a powerful tool to analytically study the training dynamics of wide neural networks. We show an interesting relationship between data augmentation and group convolutional networks. We show that equivariant NTKs outperform their non-equivariant counterparts as kernel predictors for histological image classification and quantum mechanical property prediction.
• Score: 2.373992571236766
• License:
• Abstract: Little is known about the training dynamics of equivariant neural networks, in particular how it compares to data augmented training of their non-equivariant counterparts. Recently, neural tangent kernels (NTKs) have emerged as a powerful tool to analytically study the training dynamics of wide neural networks. In this work, we take an important step towards a theoretical understanding of training dynamics of equivariant models by deriving neural tangent kernels for a broad class of equivariant architectures based on group convolutions. As a demonstration of the capabilities of our framework, we show an interesting relationship between data augmentation and group convolutional networks. Specifically, we prove that they share the same expected prediction at all training times and even off-manifold. In this sense, they have the same training dynamics. We demonstrate in numerical experiments that this still holds approximately for finite-width ensembles. By implementing equivariant NTKs for roto-translations in the plane ($G=C_{n}\ltimes\mathbb{R}^{2}$) and 3d rotations ($G=\mathrm{SO}(3)$), we show that equivariant NTKs outperform their non-equivariant counterparts as kernel predictors for histological image classification and quantum mechanical property prediction.

Related papers

• Novel Kernel Models and Exact Representor Theory for Neural Networks Beyond the Over-Parameterized Regime [52.00917519626559]
  This paper presents two models of neural-networks and their training applicable to neural networks of arbitrary width, depth and topology. We also present an exact novel representor theory for layer-wise neural network training with unregularized gradient descent in terms of a local-extrinsic neural kernel (LeNK) This representor theory gives insight into the role of higher-order statistics in neural network training and the effect of kernel evolution in neural-network kernel models.
  arXiv  Detail & Related papers  (2024-05-24T06:30:36Z)
• Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks [79.16635054977068]
  We present a new class of equivariant neural networks, dubbed Lattice-Equivariant Neural Networks (LENNs) Our approach develops within a recently introduced framework aimed at learning neural network-based surrogate models Lattice Boltzmann collision operators. Our work opens towards practical utilization of machine learning-augmented Lattice Boltzmann CFD in real-world simulations.
  arXiv  Detail & Related papers  (2024-05-22T17:23:15Z)
• Neural Tangent Kernels Motivate Graph Neural Networks with Cross-Covariance Graphs [94.44374472696272]
  We investigate NTKs and alignment in the context of graph neural networks (GNNs) Our results establish the theoretical guarantees on the optimality of the alignment for a two-layer GNN. These guarantees are characterized by the graph shift operator being a function of the cross-covariance between the input and the output data.
  arXiv  Detail & Related papers  (2023-10-16T19:54:21Z)
• Unifying O(3) Equivariant Neural Networks Design with Tensor-Network Formalism [12.008737454250463]
  We propose using fusion diagrams, a technique widely employed in simulating SU($2$)-symmetric quantum many-body problems, to design new equivariant components for equivariant neural networks. When applied to particles within a given local neighborhood, the resulting components, which we term "fusion blocks," serve as universal approximators of any continuous equivariant function. Our approach, which combines tensor networks with equivariant neural networks, suggests a potentially fruitful direction for designing more expressive equivariant neural networks.
  arXiv  Detail & Related papers  (2022-11-14T16:06:59Z)
• Interrelation of equivariant Gaussian processes and convolutional neural networks [77.34726150561087]
  Currently there exists rather promising new trend in machine leaning (ML) based on the relationship between neural networks (NN) and Gaussian processes (GP) In this work we establish a relationship between the many-channel limit for CNNs equivariant with respect to two-dimensional Euclidean group with vector-valued neuron activations and the corresponding independently introduced equivariant Gaussian processes (GP)
  arXiv  Detail & Related papers  (2022-09-17T17:02:35Z)
• Group-invariant tensor train networks for supervised learning [0.0]
  We introduce a new numerical algorithm to construct a basis of tensors that are invariant under the action of normal matrix representations. The group-invariant tensors are then combined into a group-invariant tensor train network, which can be used as a supervised machine learning model.
  arXiv  Detail & Related papers  (2022-06-30T06:33:08Z)
• Equivariant vector field network for many-body system modeling [65.22203086172019]
  Equivariant Vector Field Network (EVFN) is built on a novel equivariant basis and the associated scalarization and vectorization layers. We evaluate our method on predicting trajectories of simulated Newton mechanics systems with both full and partially observed data.
  arXiv  Detail & Related papers  (2021-10-26T14:26:25Z)
• Topographic VAEs learn Equivariant Capsules [84.33745072274942]
  We introduce the Topographic VAE: a novel method for efficiently training deep generative models with topographically organized latent variables. We show that such a model indeed learns to organize its activations according to salient characteristics such as digit class, width, and style on MNIST. We demonstrate approximate equivariance to complex transformations, expanding upon the capabilities of existing group equivariant neural networks.
  arXiv  Detail & Related papers  (2021-09-03T09:25:57Z)
• Equivariant geometric learning for digital rock physics: estimating formation factor and effective permeability tensors from Morse graph [7.355408124040856]
  We present a SE(3)-equivariant graph neural network (GNN) approach that directly predicts formation factor and effective permeability from micro-CT images. FFT solvers are established to compute both the formation factor and effective permeability, while the topology and geometry of the pore space are represented by a persistence-based Morse graph.
  arXiv  Detail & Related papers  (2021-04-12T16:28:25Z)
• Theoretical Aspects of Group Equivariant Neural Networks [9.449391486456209]
  Group equivariant neural networks have been explored in the past few years and are interesting from theoretical and practical standpoints. They leverage concepts from group representation theory, non-commutative harmonic analysis and differential geometry. In practice, they have been shown to reduce sample and model complexity, notably in challenging tasks where input transformations such as arbitrary rotations are present.
  arXiv  Detail & Related papers  (2020-04-10T17:57:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.