arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.13045cs.CVcs.GRcs.LG
收藏

PLAD:学习用伪标签和近似分布推断形状程序

PLAD: Learning to Infer Shape Programs with Pseudo-Labels and Approximate Distributions

Brown University(布朗大学);UC Berkeley(加州大学伯克利分校)

R. Kenny Jones, Homer Walke, Daniel Ritchie

摘要:

推断生成2D和3D形状的程序对于逆向工程、编辑等非常重要。执行此任务的训练模型是复杂的,因为配对(形状,程序)数据在许多领域都不容易获得,使得精确的监督学习不可行。然而,有可能通过损害分配的程序标签或形状分布的准确性来获得配对数据。唤醒-睡眠方法使用来自形状程序的生成模型的样本来近似真实形状的分布。在自我训练中,形状通过识别模型,该模型预测被视为这些形状的“伪标签”的程序。与这些方法相关,我们引入了一种新的自训练变体,该变体是程序推理所独有的,其中程序伪标签与其执行的输出形状配对,以近似形状分布为代价避免标签不匹配。我们建议将这些制度分组在一个概念框架下,其中使用来自伪标签或近似分布(PLAD)的最大似然更新进行训练。我们评估这些技术在多个2D和3D形状程序推理域。与策略梯度强化学习相比,我们表明,PLAD技术推断更准确的形状程序和收敛速度显着更快。最后,我们建议在单个模型的训练中结合来自不同PLAD方法的更新,并发现这种方法优于任何单独的技术。

英文摘要:

Inferring programs which generate 2D and 3D shapes is important for reverse engineering, editing, and more. Training models to perform this task is complicated because paired (shape, program) data is not readily available for many domains, making exact supervised learning infeasible. However, it is possible to get paired data by compromising the accuracy of either the assigned program labels or the shape distribution. Wake-sleep methods use samples from a generative model of shape programs to approximate the distribution of real shapes. In self-training, shapes are passed through a recognition model, which predicts programs that are treated as "pseudo-labels" for those shapes. Related to these approaches, we introduce a novel self-training variant unique to program inference, where program pseudo-labels are paired with their executed output shapes, avoiding label mismatch at the cost of an approximate shape distribution. We propose to group these regimes under a single conceptual framework, where training is performed with maximum likelihood updates sourced from either Pseudo-Labels or an Approximate Distribution (PLAD). We evaluate these techniques on multiple 2D and 3D shape program inference domains. Compared with policy gradient reinforcement learning, we show that PLAD techniques infer more accurate shape programs and converge significantly faster. Finally, we propose to combine updates from different PLAD methods within the training of a single model, and find that this approach outperforms any individual technique.

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。

↑