Deep learning is the mainstream technique for many machine learning tasks, including image recognition, machine translation, speech recognition, and so on. It has outperformed conventional methods in various fields and achieved great successes. Unfortunately, the understanding on how it works remains unclear. It has the central importance to lay down the theoretic foundation for deep learning. In this work, we give a geometric view to understand deep learning: we show that the fundamental principle attributing to the success is the manifold structure in data, namely natural high dimensional data concentrates close to a low-dimensional manifold, deep learning learns the manifold and the probability distribution on it. We further introduce the concepts of rectified linear complexity for deep neural network measuring its learning capability, rectified linear complexity of an embedding manifold describing the difficulty to be learned. Then we show for any deep neural network with fixed architecture, there exists a manifold that cannot be learned by the network. By empirical evidences, we also demonstrate the learning accuracies of the-state-of-art autoencoders are reasonably good but still leave large spaces to be improved. Finally, we propose to apply optimal mass transportation theory to control the probability distribution in the latent space.

该研究通过几何视角揭示深度学习的关键在于学习数据的流形结构和概率分布，并提出一种控制潜在空间中概率分布的最优质量传递理论。

深度学习的几何理解