arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2206.04833cs.LG
收藏

使用SAT求解器训练神经网络

Training Neural Networks using SAT solvers

Indian Institute of Technology - Kharagpur(印度理工学院卡拉格布尔分校)

Subham S. Sahoo

摘要:

我们提出了一种算法,探索全局优化方法,使用SAT求解器,训练神经网络。深度神经网络在图像识别、语音识别等任务中取得了巨大的成就,它们的成功在很大程度上归功于基于梯度的优化方法,这种方法可以很好地扩展到巨大的数据集,同时仍然提供比任何其他现有方法更好的解决方案。然而,存在学习问题,如奇偶函数和快速傅立叶变换,其中使用基于梯度的优化算法的神经网络不能正确地捕获学习任务的底层结构。因此,探索全局优化方法是最感兴趣的,因为基于梯度的方法陷入局部最优。在实验中,我们证明了我们的算法在某些任务中(如奇偶校验学习)对ADAM优化器的有效性。然而,在图像分类的MNIST数据集的情况下,我们的算法的性能是不太令人满意的。我们进一步讨论了训练数据集的大小和超参数设置在保持SAT求解器的可扩展性方面的作用。

英文摘要:

We propose an algorithm to explore the global optimization method, using SAT solvers, for training a neural net. Deep Neural Networks have achieved great feats in tasks like-image recognition, speech recognition, etc. Much of their success can be attributed to the gradient-based optimisation methods, which scale well to huge datasets while still giving solutions, better than any other existing methods. However, there exist learning problems like the parity function and the Fast Fourier Transform, where a neural network using gradient-based optimisation algorithm can not capture the underlying structure of the learning task properly. Thus, exploring global optimisation methods is of utmost interest as the gradient-based methods get stuck in local optima. In the experiments, we demonstrate the effectiveness of our algorithm against the ADAM optimiser in certain tasks like parity learning. However, in the case of image classification on the MNIST Dataset, the performance of our algorithm was less than satisfactory. We further discuss the role of the size of the training dataset and the hyper-parameter settings in keeping things scalable for a SAT solver.

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。

↑