Towards a Scalable Reference-Free Evaluation of Generative Models
This work addresses a computational bottleneck for researchers and practitioners evaluating large-scale generative models, though it is incremental as it builds on existing entropy scores.
The authors tackled the computational inefficiency of reference-free entropy scores (VENDI and RKE) for evaluating generative model diversity by proposing the Fourier-based Kernel Entropy Approximation (FKEA) method, which reduces complexity to O(n) and demonstrates scalability and interpretability on image, text, and video datasets.
While standard evaluation scores for generative models are mostly reference-based, a reference-dependent assessment of generative models could be generally difficult due to the unavailability of applicable reference datasets. Recently, the reference-free entropy scores, VENDI and RKE, have been proposed to evaluate the diversity of generated data. However, estimating these scores from data leads to significant computational costs for large-scale generative models. In this work, we leverage the random Fourier features framework to reduce the computational price and propose the Fourier-based Kernel Entropy Approximation (FKEA) method. We utilize FKEA's approximated eigenspectrum of the kernel matrix to efficiently estimate the mentioned entropy scores. Furthermore, we show the application of FKEA's proxy eigenvectors to reveal the method's identified modes in evaluating the diversity of produced samples. We provide a stochastic implementation of the FKEA assessment algorithm with a complexity $O(n)$ linearly growing with sample size $n$. We extensively evaluate FKEA's numerical performance in application to standard image, text, and video datasets. Our empirical results indicate the method's scalability and interpretability applied to large-scale generative models. The codebase is available at https://github.com/aziksh-ospanov/FKEA.