Statistical learning by sparse deep neural networks
arXiv:2311.08845v11 citationsh-index: 22
Originality Incremental advance
AI Analysis
This provides theoretical foundations for sparse deep learning, addressing a core statistical challenge in machine learning.
The authors tackled the problem of deriving statistical guarantees for deep neural networks with l1-regularization, proving that the estimator achieves adaptively near-minimax excess risk across multiple function classes in regression and classification.
We consider a deep neural network estimator based on empirical risk minimization with l_1-regularization. We derive a general bound for its excess risk in regression and classification (including multiclass), and prove that it is adaptively nearly-minimax (up to log-factors) simultaneously across the entire range of various function classes.