Guardado en:
Detalles Bibliográficos
Autor principal: Raj, Prithvi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2506.14167
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909030331449344
author Raj, Prithvi
author_facet Raj, Prithvi
contents Generative models typically rely on either simple latent priors (e.g., Variational Autoencoders, VAEs), which are efficient but limited, or highly expressive iterative samplers (e.g., Diffusion and Energy-based Models), which are costly and opaque. We introduce the Kolmogorov-Arnold Energy Model (KAEM) to bridge this trade-off and provide new opportunities for latent-space interpretability. Based on a novel adaptation of the Kolmogorov-Arnold Representation Theorem, KAEM imposes a univariate latent structure on the prior, enabling exact inference via the inverse transform method. With a low-dimensional latent space and appropriate inductive biases, importance sampling becomes a tractable, unbiased, and efficient posterior inference method. For settings where this fails, we propose a population-based strategy that decomposes the posterior into a sequence of annealed distributions, a new remedy for poor mixing in Energy-based Models. We compare KAEM against VAEs and the neural latent EBM architecture. KAEM attains the best Fréchet Inception Distance among latent-prior models on SVHN and CIFAR10, while sampling in a single forward pass and exposing an interpretable prior built from 1D densities.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14167
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Kolmogorov-Arnold Energy Models: Fast, Interpretable Generative Modeling
Raj, Prithvi
Machine Learning
Generative models typically rely on either simple latent priors (e.g., Variational Autoencoders, VAEs), which are efficient but limited, or highly expressive iterative samplers (e.g., Diffusion and Energy-based Models), which are costly and opaque. We introduce the Kolmogorov-Arnold Energy Model (KAEM) to bridge this trade-off and provide new opportunities for latent-space interpretability. Based on a novel adaptation of the Kolmogorov-Arnold Representation Theorem, KAEM imposes a univariate latent structure on the prior, enabling exact inference via the inverse transform method. With a low-dimensional latent space and appropriate inductive biases, importance sampling becomes a tractable, unbiased, and efficient posterior inference method. For settings where this fails, we propose a population-based strategy that decomposes the posterior into a sequence of annealed distributions, a new remedy for poor mixing in Energy-based Models. We compare KAEM against VAEs and the neural latent EBM architecture. KAEM attains the best Fréchet Inception Distance among latent-prior models on SVHN and CIFAR10, while sampling in a single forward pass and exposing an interpretable prior built from 1D densities.
title Kolmogorov-Arnold Energy Models: Fast, Interpretable Generative Modeling
topic Machine Learning
url https://arxiv.org/abs/2506.14167