Two Facets of SDE Under an Information-Theoretic Lens: Generalization of SGD via Training Trajectories and via Terminal States
원문 초록
Stochastic differential equations (SDEs) have been shown recently to characterize well the dynamics of training machine learning models with SGD. When the generalization error of the SDE approximation closely aligns with that of SGD in expectation, it provides two opportunities for understanding better the generalization behaviour of SGD through its SDE approximation. Firstly, viewing SGD as full-batch gradient descent with Gaussian gradient noise allows us to obtain trajectory-based generalization bound using the information-theoretic bound from Xu and Raginsky [2017]. Secondly, assuming mild conditions, we estimate the steady-state weight distribution of SDE and use information-theoretic bounds from Xu and Raginsky [2017] and Negrea et al. [2019] to establish terminal-state-based generalization bounds. Our proposed bounds have some advantages, notably the trajectory-based bound outperforms results in Wang and Mao [2022], and the terminal-state-based bound exhibits a fast decay rate comparable to stability-based bounds.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Learning Non-Vacuous Generalization Bounds from Optimization
Learning Where to Learn: Training Distribution Selection for Provable OOD Performance
Operator Learning for Schrödinger Equation: Unitarity, Error Bounds, and Time Generalization
Graph Representational Learning: When Does More Expressivity Hurt Generalization?
Model Steering: Learning with a Reference Model Improves Generalization Bounds and Scaling Laws
Achieving Distributive Justice in Federated Learning via Uncertainty Quantification
토론
이 논문과 연결된 포럼 글이 아직 없습니다.