Current methods for estimating the required neural-network size for a given problem class have focused on methods that can be computationally intensive, such as neural-architecture search and pruning. In contrast, methods that add capacity to neural networks as needed may provide similar results to architecture search and pruning, but do not require as much computation to find an appropriate network size. Here, we present a network-growth method that searches for explainable error in the network's residuals and grows the network if sufficient error is detected. We demonstrate this method using examples from classification, imitation learning, and reinforcement learning. Within these tasks, the growing network can often achieve better performance than small networks that do not grow, and similar performance to networks that begin much larger.
翻译:目前用于估计特定问题类所需的神经网络规模的方法侧重于可以计算密集的方法,例如神经结构搜索和修剪。相比之下,增加神经网络能力的方法可能为结构搜索和修剪提供类似的结果,但不需要多少计算来找到合适的网络规模。在这里,我们提出了一个网络增长方法,在发现足够的错误时,搜索网络剩余部分中可解释的错误,并发展网络。我们用分类、模仿学习和强化学习等实例来显示这一方法。在这些任务中,不断增长的网络往往比没有增长的小网络取得更好的性能,与开始大得多的网络的类似性能。