ProtoVAE：用于无监督解耦的原型网络
ProtoVAE: Prototypical Networks for Unsupervised Disentanglement
AI总结：
本文提出基于深度生成式VAE的无监督解耦模型ProtoVAE，利用自监督训练的深度度量学习原型网络施加约束，确保表示空间的受控变化映射到数据空间的变化因子，在多个基准数据集上取得了最先进结果。
摘要：
生成式建模和自监督学习近年来在完全无监督地从数据中学习方面取得了巨大进展。然而，如何引导神经网络将数据编码为可解释或可说明的表示，仍然是一个开放的研究领域。无监督解耦问题尤为重要，因为它提出仅从数据中发现不同的潜在变化因子或语义概念，无需标记样本，并将它们编码为结构上互不交叠的潜在表示。如果在网络中没有加入额外的约束或归纳偏置，生成模型可能会学习到数据分布并编码这些因子，但不一定是解耦的方式。在此，我们引入了一种新颖的基于深度生成式 VAE 的模型 ProtoVAE，该模型利用通过自监督训练的深度度量学习 Prototypical network（原型网络）来施加这些约束。该原型网络约束了从表示空间到数据空间的映射，以确保表示空间中的受控变化能映射到数据空间中变化因子的改变。我们的模型完全无监督，不需要关于数据集的任何先验知识，包括因子的数量。我们在基准数据集 dSprites、3DShapes 和 MPI3D 解耦数据集上评估了我们提出的模型，通过潜在空间中的定性遍历以及定量解耦指标，展示了相对于以往方法的最先进结果。我们进一步在真实世界的 CelebA 数据集上定性展示了我们模型的有效性。
英文摘要：
Generative modeling and self-supervised learning have in recent years made great strides towards learning from data in a completely unsupervised way. There is still however an open area of investigation into guiding a neural network to encode the data into representations that are interpretable or explainable. The problem of unsupervised disentanglement is of particular importance as it proposes to discover the different latent factors of variation or semantic concepts from the data alone, without labeled examples, and encode them into structurally disjoint latent representations. Without additional constraints or inductive biases placed in the network, a generative model may learn the data distribution and encode the factors, but not necessarily in a disentangled way. Here, we introduce a novel deep generative VAE-based model, ProtoVAE, that leverages a deep metric learning Prototypical network trained using self-supervision to impose these constraints. The prototypical network constrains the mapping of the representation space to data space to ensure that controlled changes in the representation space are mapped to changes in the factors of variations in the data space. Our model is completely unsupervised and requires no a priori knowledge of the dataset, including the number of factors. We evaluate our proposed model on the benchmark dSprites, 3DShapes, and MPI3D disentanglement datasets, showing state of the art results against previous methods via qualitative traversals in the latent space, as well as quantitative disentanglement metrics. We further qualitatively demonstrate the effectiveness of our model on the real-world CelebA dataset.
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。
