论文标题
并非所有彩票都平等
Not All Lotteries Are Made Equal
论文作者
论文摘要
彩票票证假设（LTH）指出，对于合理尺寸的神经网络，同一网络中的子网络的性能与从相同初始化进行训练时的密集性相比，其性能不低。这项工作调查了模型大小与查找这些稀疏子网络的易用性之间的关系。我们通过实验表明，令人惊讶的是，在有限的预算下，较小的型号从票证搜索（TS）中受益更多。
The Lottery Ticket Hypothesis (LTH) states that for a reasonably sized neural network, a sub-network within the same network yields no less performance than the dense counterpart when trained from the same initialization. This work investigates the relation between model size and the ease of finding these sparse sub-networks. We show through experiments that, surprisingly, under a finite budget, smaller models benefit more from Ticket Search (TS).
