Exploring explicit coarse-grained structure in artificial neural networks

Abstract

We propose to employ the hierarchical coarse-grained structure in the artificial neural networks explicitly to improve the interpretability without degrading performance. The idea has been applied in two situations. One is a neural network called TaylorNet, which aims to approximate the general mapping from input data to output result in terms of Taylor series directly, without resorting to any magic nonlinear activations. The other is a new setup for data distillation, which can perform multi-level abstraction of the input dataset and generate new data that possesses the relevant features of the original dataset and can be used as references for classification. In both cases, the coarse-grained structure plays an important role in simplifying the network and improving both the interpretability and efficiency. The validity has been demonstrated on MNIST and CIFAR-10 datasets. Further improvement and some open questions related are also discussed.

References in corpus (10)

• Exact evolution equation for the effective potential
• MLP-Mixer: An all-MLP Architecture for Vision
• Matrix Product Density Operators: Simulation of finite-T and dissipative systems
• Learning phase transitions by confusion
• Tree Tensor Networks for Generative Modeling
• Variational Neural Annealing
• Tensor lattice field theory with applications to the renormalization group and quantum computing
• Supervised Learning with Projected Entangled Pair States
• Deep Learning the Functional Renormalization Group
• Deep tensor networks with matrix product operators