arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.12063cs.LGstat.ML
收藏

深度残差混合模型

Deep Residual Mixture Models

Aalto University(阿尔托大学)

Perttu Hämäläinen, Martin Trapp, Tuure Saloheimo, Arno Solin

摘要:

我们提出了深度残差混合模型(DRMOST),一种新的深度生成模型架构。与其他深度模型相比,DRNN允许更灵活的条件采样:模型可以用所有变量训练一次,然后用于条件变量、高斯先验和(非)等式约束的任意组合的采样。这为交互式和探索性机器学习提供了新的机会,其中应该最大限度地减少用户等待重新训练模型的时间。我们证明了DRDO在约束多肢体逆运动学和可控的动画生成。

英文摘要:

We propose Deep Residual Mixture Models (DRMMs), a novel deep generative model architecture. Compared to other deep models, DRMMs allow more flexible conditional sampling: The model can be trained once with all variables, and then used for sampling with arbitrary combinations of conditioning variables, Gaussian priors, and (in)equality constraints. This provides new opportunities for interactive and exploratory machine learning, where one should minimize the user waiting for retraining a model. We demonstrate DRMMs in constrained multi-limb inverse kinematics and controllable generation of animations.

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。

↑