Sparse Deep Learning: A New Framework Immune to Local Traps and Miscalibration
arXiv:2110.00653
Abstract
Deep learning has powered recent successes of artificial intelligence (AI). However, the deep neural network, as the basic model of deep learning, has suffered from issues such as local traps and miscalibration. In this paper, we provide a new framework for sparse deep learning, which has the above issues addressed in a coherent way. In particular, we lay down a theoretical foundation for sparse deep learning and propose prior annealing algorithms for learning sparse neural networks. The former has successfully tamed the sparse deep neural network into the framework of statistical modeling, enabling prediction uncertainty correctly quantified. The latter can be asymptotically guaranteed to converge to the global optimum, enabling the validity of the down-stream statistical inference. Numerical result indicates the superiority of the proposed method compared to the existing ones.
Neurips 2021
References in corpus (9)
- A Convergence Theory for Deep Learning via Over-Parameterization
- Stochastic Gradient Hamiltonian Monte Carlo
- Sparse Networks from Scratch: Faster Training without Losing Performance
- Variable selection for BART: An application to gene regulation
- On the Convergence of Stochastic Gradient MCMC Algorithms with High-Order Integrators
- Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterization
- Dynamic Model Pruning with Feedback
- Sparse-Input Neural Networks for High-dimensional Nonparametric Regression and Classification
- Posterior Concentration for Sparse Deep Learning