Properties of interest for crystals and molecules, such as band gap, elasticity, and solubility, are generally related to each other: they are governed by the same underlying laws of physics. However, when state-of-the-art graph neural networks attempt to predict multiple properties simultaneously (the multi-task learning (MTL) setting), they frequently underperform a suite of single property predictors. This suggests graph networks may not be fully leveraging these underlying similarities. Here we investigate a potential explanation for this phenomenon: the curvature of each property's loss surface significantly varies, leading to inefficient learning. This difference in curvature can be assessed by looking at spectral properties of the Hessians of each property's loss function, which is done in a matrix-free manner via randomized numerical linear algebra. We evaluate our hypothesis on two benchmark datasets (Materials Project (MP) and QM8) and consider how these findings can inform the training of novel multi-task learning models.
翻译:晶体和分子的兴趣属性,例如波段间距、弹性和溶解性,一般是相互联系的:它们受同样的物理基本规律的制约。然而,当最先进的图形神经网络试图同时预测多种属性(多任务学习(MTL)设置)时,它们往往低于一组单一属性预测仪。这表示图形网络可能无法充分利用这些内在相似性。我们在这里调查了这一现象的潜在解释:每个地产损失表面的曲线差异很大,导致学习效率低下。可以通过查看每种地产损失函数的赫斯人光谱特性来评估这种曲线差异,而这种光谱特性是通过随机数字线性代数以不设矩阵的方式完成的。我们评估了我们关于两个基准数据集(马斯特列工程(MP)和QM8)的假设,并考虑这些发现如何为新型多任务学习模型的培训提供信息。