将GANs解释为部分随机的贝叶斯神经网络，揭示其成功与局限性。
Bridging GANs and Bayesian Neural Networks via Partial Stochasticity
- 把GANs看作部分随机的贝叶斯网络，统一理解其优化机制
- 提出平滑损失曲面和最小描述长度搜索策略，提升性能
- 适合研究生成模型理论或优化方法的读者
生成对抗网络（GANs）是流行且成功的生成模型，但其优化过程极其困难。本文通过将GANs视为具有部分随机性的贝叶斯神经网络，解释了其成功与局限性。这一视角使我们得以建立通用近似条件，并将多种GAN变体的对抗式优化转化为对边缘化随机变量后得到的似然代理函数的优化。基于此解释，正则化需求变得明显，因此我们提出平滑损失曲面的策略以及寻找最小描述长度解的方法，这些方法与平坦极小值相关，有助于良好泛化。在广泛实验中，这些策略显著提升了性能，为深入理解GANs开辟了新路径。
原文摘要 · Abstract (English)
Generative Adversarial Networks (GANs) are popular and successful generative models. Despite their success, optimization is notoriously challenging. In this work, we explain the success and limitations of GANs by casting them as Bayesian neural networks with partial stochasticity. This interpretation allows us to establish conditions of universal approximation and to rewrite the adversarial-style optimization of several variants of GANs as the optimization of a proxy for the likelihood obtained by marginalizing out the stochastic variables. Following this interpretation, the need for regularization becomes apparent, and we propose to adopt strategies to smooth the loss landscape and methods to search for solutions with minimum description length, which are associated with flat minima and good generalization. Results obtained on a wide range of experiments indicate that these strategies lead to performance improvements and pave the way to a deeper understanding of GANs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。