按年被提到
20202021 · 2 次2026
视觉、语音与图像生成›视觉语言与自监督表征·arXiv:2011.10650 · cs.LG

Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images

Rewon Child · 2020
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2021-01-20 · 最近 2021-05-12
2不同的人
2活跃月份
2讨论串
171被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We present a hierarchical VAE that, for the first time, generates samples quickly while outperforming the PixelCNN in log-likelihood on all natural image benchmarks. We begin by observing that, in theory, VAEs can actually represent autoregressive models, as well as faster, better models if they exist, when made sufficiently deep. Despite this, autoregressive models have historically outperformed VAEs in log-likelihood. We test if insufficient depth explains why by scaling a VAE to greater stochastic depth than previously explored and evaluating it CIFAR-10, ImageNet, and FFHQ. In comparison to the PixelCNN, these very deep VAEs achieve higher likelihoods, use fewer parameters, generate samples thousands of times faster, and are more easily applied to high-resolution images. Qualitative studies suggest this is because the VAE learns efficient hierarchical visual representations. We release our source code and models at https://github.com/openai/vdvae.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论gwern · 2021-01

“If you don't have TFRC access, this is out of the question as a hobbyist. It's >32 GPU-months. Sorry. However, you can still transfer learn this particular model with <1 GPU-month, if you have some anime or artwork dataset you want (eg a specific character). The good news is that there's been huge progress in GAN efficiency over the past few years, and w…”

This Anime Does Not Exist原评论 ↗
评论ansk · 2021-05

“For an actual example of this more cautious framing, see another paper (also by OpenAI, also comparing two classes of generative models), https://arxiv.org/abs/2011.10650.”

Diffusion Models Beat GANs on Image Synthesis原评论 ↗