Training Diffusion Models with Reinforcement Learning,arXiv - CS - Computer Vision and Pattern Recognition

相关文章推荐

千年单身的蚂蚁 · “镍王”的36岁女婿总裁，能扛起储能的大旗吗？· 4 月前 ·

狂野的啄木鸟 · 超新星全运会第二季_百度百科· 4 月前 ·

重情义的炒粉 · 大竹县人民医院_百度百科· 5 月前 ·

忐忑的黑框眼镜 · 中国南极科考队内陆队抵达泰山站-新华网· 1 年前 ·

打酱油的柿子 · 湖北文理学院理工学院2023年招聘公告-高校人才网· 2 年前 ·

Diffusion models are a class of flexible generative models trained with an approximation to the log-likelihood objective. However, most use cases of diffusion models are not concerned with likelihoods, but instead with downstream objectives such as human-perceived image quality or drug effectiveness. In this paper, we investigate reinforcement learning methods for directly optimizing diffusion models for such objectives. We describe how posing denoising as a multi-step decision-making problem enables a class of policy gradient algorithms, which we refer to as denoising diffusion policy optimization (DDPO), that are more effective than alternative reward-weighted likelihood approaches. Empirically, DDPO is able to adapt text-to-image diffusion models to objectives that are difficult to express via prompting, such as image compressibility, and those derived from human feedback, such as aesthetic quality. Finally, we show that DDPO can improve prompt-image alignment using feedback from a vision-language model without the need for additional data collection or human annotation. 中文翻译：

推荐文章

千年单身的蚂蚁 · “镍王”的36岁女婿总裁，能扛起储能的大旗吗？

4 月前

狂野的啄木鸟 · 超新星全运会第二季_百度百科

4 月前

重情义的炒粉 · 大竹县人民医院_百度百科

5 月前

忐忑的黑框眼镜 · 中国南极科考队内陆队抵达泰山站-新华网

1 年前

打酱油的柿子 · 湖北文理学院理工学院2023年招聘公告-高校人才网

2 年前