Every Local Minimum Value is the Global Minimum Value of Induced Model in Non-convex Machine Learning
This provides a foundational theoretical guarantee for nonconvex optimization in machine learning, applicable to models like deep neural networks without modifying practical methods.
The paper proves that in nonconvex machine learning, every local minimum achieves the global optimum of a perturbable gradient basis model, making nonconvex optimization theoretically as supported as convex optimization with handcrafted bases, except when preferring handcrafted bases.
For nonconvex optimization in machine learning, this article proves that every local minimum achieves the globally optimal value of the perturbable gradient basis model at any differentiable point. As a result, nonconvex machine learning is theoretically as supported as convex machine learning with a handcrafted basis in terms of the loss at differentiable local minima, except in the case when a preference is given to the handcrafted basis over the perturbable gradient basis. The proofs of these results are derived under mild assumptions. Accordingly, the proven results are directly applicable to many machine learning models, including practical deep neural networks, without any modification of practical methods. Furthermore, as special cases of our general results, this article improves or complements several state-of-the-art theoretical results on deep neural networks, deep residual networks, and overparameterized deep neural networks with a unified proof technique and novel geometric insights. A special case of our results also contributes to the theoretical foundation of representation learning.