论文标题
通过放松的注射概率流进行正则自动编码器
Regularized Autoencoders via Relaxed Injective Probability Flow
论文作者
论文摘要
基于流动的生成模型是学习生成样品的有效方法，同时允许进行可拖动的可能性计算和推理。但是，可逆性要求限制了模型具有与输入相同的潜在维度。这构成了巨大的建筑，内存和计算成本，使其比其他类别的生成模型（例如变分自动编码器（VAE））更具挑战性。我们提出了一个基于概率流的生成模型，该模型消除了模型上的射击性要求，并且仅假定注射率。这也提供了对正规自动编码器（RAE）的另一种观点，我们的最终目标类似于RA与特定的正规化器，这些正则化是通过下降概率流目标来得出的。我们从经验上证明了所提出的模型的希望，从样本质量方面改善了VAE和AES。
Invertible flow-based generative models are an effective method for learning to generate samples, while allowing for tractable likelihood computation and inference. However, the invertibility requirement restricts models to have the same latent dimensionality as the inputs. This imposes significant architectural, memory, and computational costs, making them more challenging to scale than other classes of generative models such as Variational Autoencoders (VAEs). We propose a generative model based on probability flows that does away with the bijectivity requirement on the model and only assumes injectivity. This also provides another perspective on regularized autoencoders (RAEs), with our final objectives resembling RAEs with specific regularizers that are derived by lower bounding the probability flow objective. We empirically demonstrate the promise of the proposed model, improving over VAEs and AEs in terms of sample quality.
