Neural Functional Transformers

Published: 21 Sept 2023, Last Modified: 31 Jan 2024NeurIPS 2023 posterEveryoneRevisionsBibTeX
Keywords: equivariance, permutation, implicit neural representation, generalization, transformers, attention
TL;DR: Neural Functional Transformers (NFTs) are deep permutation-equivariant weight-space models based on attention mechanisms, enabling greater expressivity and improved performance in tasks such as learning latent representation of weights.
Abstract: The recent success of neural networks as implicit representation of data has driven growing interest in neural functionals: models that can process other neural networks as input by operating directly over their weight spaces. Nevertheless, constructing expressive and efficient neural functional architectures that can handle high-dimensional weight-space objects remains challenging. This paper uses the attention mechanism to define a novel set of permutation equivariant weight-space layers and composes them into deep equivariant models called neural functional Transformers (NFTs). NFTs respect weight-space permutation symmetries while incorporating the advantages of attention, which have exhibited remarkable success across multiple domains. In experiments processing the weights of feedforward MLPs and CNNs, we find that NFTs match or exceed the performance of prior weight-space methods. We also leverage NFTs to develop Inr2Array, a novel method for computing permutation invariant latent representations from the weights of implicit neural representations (INRs). Our proposed method improves INR classification accuracy by up to $+17\\%$ over existing methods. We provide an implementation of our layers at https://github.com/AllanYangZhou/nfn.
Submission Number: 2577
Loading