Research map / Optimization / Nonconvex and smooth optimization
Black-Box Generalization: Stability of Zeroth-Order Learning
Konstantinos Nikolakakis, Farzin Haddadpour, Dionysios Kalogerias, Amin Karbasi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace gradient updates with ZoSS loss-difference updates and decrease the step size by Γ_K^d | SGD-style learning under Lipschitz, smooth losses · Bounded nonconvex loss | generalization error bound: \dfrac{1+(CB)^{-1}}{n}\big(2CL^{2}\big)^{\frac{1}{C\beta+1}}(eT)^{\frac{C\beta}{C\beta+1}} -> \dfrac{1+(CB)^{-1}}{n}\big | Mixed |
| Use α_t≤C/t instead of a step size scaled by Γ_K^d | ZoSS · Bounded nonconvex loss, single-example updates | generalization error bound: The bound is \dfrac{3e\big(1+(CB)^{-1}\big)^{2}}{2n}\big(1+(2+c)CL^{2}\big)T and is independent of d and K. | Improved |
| Increase the number of loss evaluations K | ZoSS · High-dimensional black-box learning | generalization error | Mixed |
| Use mini-batch updates rather than single-example updates | ZoSS · Unbounded loss, any batch size m≤n | generalization error bound | No clear effect |
| Take K→∞ and c→0 | ZoSS · Generalization bounds | generalization error bound | Improved |
| Use a smaller smoothing parameter μ satisfying the stated stability condition | ZoSS · Lipschitz, smooth losses | generalization error bound | Improved |
Closest papers by meaning
- Towards Generalization beyond Pointwise Learning: A Unified Information-theoretic Perspective (ICML 2024 · similarity 0.88)
- Towards Stability and Generalization Bounds in Decentralized Minibatch Stochastic Gradient Descent (AAAI 2024 · similarity 0.88)
- Zeroth-Order Optimization Finds Flat Minima (NeurIPS 2025 · similarity 0.88)
- Generalizability of Neural Networks Minimizing Empirical Risk Based on Expressive Power (ICLR 2025 · similarity 0.88)
- Generalization Bound of Gradient Flow through Training Trajectory and Data-dependent Kernel (NeurIPS 2025 · similarity 0.88)
- Algorithmic Stability Unleashed: Generalization Bounds with Unbounded Losses (ICML 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
