去噪多Beta VAE：用于解耦与生成的表示学习
Denoising Multi-Beta VAE: Representation Learning for Disentanglement and Generation
AI总结：
提出去噪多Beta VAE框架，通过训练单一VAE获得多β值表示，并利用非线性扩散模型平滑过渡，平衡解耦与生成质量，支持独立样本生成。
摘要：
在生成模型中，解耦且可解释的潜在表示通常以牺牲生成质量为代价。$\eta$-VAE框架引入超参数$\eta$来平衡解耦与重建质量，其中设置$\eta>1$会引入信息瓶颈，偏向于解耦而非清晰、准确的重建。为解决这一权衡，我们提出了一种新颖的生成建模框架，利用一系列$\eta$值来学习多个对应的潜在表示。首先，我们通过训练单个变分自编码器（VAE）获得大量表示，使用一种新的损失函数来控制每个潜在表示中保留的信息，使得较高的$\eta$值优先考虑解耦而非重建保真度。然后，我们引入一个非线性扩散模型，平滑地过渡对应于不同$\eta$值的潜在表示。该模型向较少解耦且更具信息性的表示去噪，最终导致（几乎）无损的表示，从而实现清晰的重建。此外，我们的模型支持无需输入图像的样本生成，可作为独立的生成模型运行。我们在解耦和生成质量两方面评估了我们的框架。另外，我们观察到潜在空间中随$\eta$变化的平滑过渡，便于对生成输出进行一致的操作。
英文摘要：
Disentangled and interpretable latent representations in generative models typically come at the cost of generation quality. The $β$-VAE framework introduces a hyperparameter $β$ to balance disentanglement and reconstruction quality, where setting $β> 1$ introduces an information bottleneck that favors disentanglement over sharp, accurate reconstructions. To address this trade-off, we propose a novel generative modeling framework that leverages a range of $β$ values to learn multiple corresponding latent representations. First, we obtain a slew of representations by training a single variational autoencoder (VAE), with a new loss function that controls the information retained in each latent representation such that the higher $β$ value prioritize disentanglement over reconstruction fidelity. We then, introduce a non-linear diffusion model that smoothly transitions latent representations corresponding to different $β$ values. This model denoises towards less disentangled and more informative representations, ultimately leading to (almost) lossless representations, enabling sharp reconstructions. Furthermore, our model supports sample generation without input images, functioning as a standalone generative model. We evaluate our framework in terms of both disentanglement and generation quality. Additionally, we observe smooth transitions in the latent spaces with respect to changes in $β$, facilitating consistent manipulation of generated outputs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
