Machine learning systems typically apply the same model to both easy and tough cases. This is in stark contrast with humans, who tend to evoke either fast (instinctive) or slow (analytical) thinking depending on the problem difficulty, a property called the dual-process theory of mind. We present FLOWGEN, a graph-generation model inspired by the dual-process theory of mind that generates large graphs incrementally. Depending on the difficulty of completing the graph at the current step, graph generation is routed to either a fast (weaker) or a slow (stronger) model. The fast and slow models have identical architectures, but vary in the number of parameters and consequently the strength. Experiments on real-world graphs show that ours can successfully generate graphs similar to those generated by a single large model in a fraction of time.
翻译:机器学习系统通常对容易和困难的情况都采用同样的模型。这与人类形成鲜明对比,人类往往根据问题的困难而快速(诱导性)或缓慢(分析性)思考(分析性),这是一个称为双过程思维理论的属性。我们介绍了FLOWGEN,这是一个由双过程思维理论所启发的图形生成模型,它会逐渐生成大图。根据在目前阶段完成图表的困难,图形生成被选择为快速(较弱)或慢(较强)模型。快速和缓慢的模型具有相同的结构,但参数的数量和强度各不相同。在现实世界的图表上进行的实验显示,我们能够成功地生成与一个单一大模型在很短的时间里生成的图形相似的图形。