改进的硬注意神经网络训练中的近优扫视序列
Near-Optimal Glimpse Sequences for Improved Hard Attention Neural Network Training
摘要:
硬视觉注意是减少现代计算机视觉方法计算负担的一种很有前途的方法。硬注意力机制通常是不可微的。它们可以通过强化学习进行训练,但这需要高方差的训练,这阻碍了更广泛的应用。我们展示了如何努力注意图像分类可以作为一个贝叶斯最优实验设计(BOED)的问题。从这个角度来看,要注意的最佳位置是那些提供分类分布的熵的最大预期减少的位置。我们介绍的方法从BOED文献近似这种最佳的行为,并使用它来产生“近最佳”序列的注意力位置。然后,我们将展示如何使用这样的序列部分监督,因此加快,一个硬注意力机制的训练。虽然生成这些序列的计算成本很高,但它们可以被任何其他网络重用,这些网络后来在同一任务上进行了训练。
英文摘要:
Hard visual attention is a promising approach to reduce the computational burden of modern computer vision methodologies. Hard attention mechanisms are typically non-differentiable. They can be trained with reinforcement learning but the high-variance training this entails hinders more widespread application. We show how hard attention for image classification can be framed as a Bayesian optimal experimental design (BOED) problem. From this perspective, the optimal locations to attend to are those which provide the greatest expected reduction in the entropy of the classification distribution. We introduce methodology from the BOED literature to approximate this optimal behaviour, and use it to generate `near-optimal' sequences of attention locations. We then show how to use such sequences to partially supervise, and therefore speed up, the training of a hard attention mechanism. Although generating these sequences is computationally expensive, they can be reused by any other networks later trained on the same task.
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。