MIND: Monge Inception Distance for Generative Models Evaluation
For researchers and practitioners evaluating generative models, MIND provides a more efficient and robust alternative to FID, enabling faster iteration and reliable evaluation with smaller sample sizes.
The paper proposes MIND, a metric for evaluating generative models that uses sliced Wasserstein distance to overcome FID's limitations. MIND achieves one order of magnitude better sample efficiency, two orders of magnitude faster computation, and greater robustness to adversarial attacks, matching FID's evaluation performance with 5k samples instead of 50k.
We propose the Monge Inception Distance (MIND), a metric for evaluating generative models that addresses key limitations of the widely adopted Fréchet Inception Distance (FID). The MIND metric leverages the sliced Wasserstein distance to compare distributions by averaging one-dimensional optimal transport distances, efficiently computed via sorting. This approach circumvents the estimation of high-dimensional means and covariance matrices, which underlie FID's poor sample complexity and vulnerability to adversarial attacks. We empirically demonstrate three primary advantages: (i) it is more sample-efficient by one order of magnitude, (ii) it is faster to compute by two orders of magnitude, (iii) it is more robust to adversarial attacks such as moment-matching. We show that MIND with 5k samples can replace the evaluation performance of FID with 50k samples, providing high correlation with this standard benchmark and superior discriminative performance. We further demonstrate that even smaller sample sizes (e.g., 1k or 2k) remain highly informative for rapid model iteration.