Optimization of Smooth Functions with Noisy Observations: Local Minimax Rates

Authors: Yining Wang, Sivaraman Balakrishnan, Aarti Singh

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We propose a local minimax framework to study the fundamental difficulty of optimizing smooth functions with adaptive function evaluations. Our main results are to characterize the local convergence rates Rnpf0q for a wide range of reference functions f0 P F. We prove local minimax lower bounds that match the n α{p2α d αβq upper bound, up to logarithmic factors in n.
Researcher Affiliation Academia Yining Wang, Sivaraman Balakrishnan, Aarti Singh Department of Machine Learning and Statistics Carnegie Mellon University, Pittsburgh, PA, 15213, USA {yiningwa,aarti}@cs.cmu.edu, siva@stat.cmu.edu
Pseudocode No The paper includes a conceptual illustration of an algorithm in Figure 1 and describes its procedure in Section A. However, it does not present a formal, structured pseudocode block labeled "Algorithm" or "Pseudocode".
Open Source Code No The paper does not provide any statement about releasing source code or a link to a code repository for the methodology described.
Open Datasets No The paper is theoretical and models a system (Eq. 1) but does not use real-world datasets for training or experimentation. Therefore, no information about publicly available datasets is provided.
Dataset Splits No The paper is theoretical and does not involve empirical experiments with datasets. Therefore, no information regarding training, validation, or test splits is provided.
Hardware Specification No The paper is theoretical and does not describe any computational experiments. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not mention any specific software components or their versions required for reproduction.
Experiment Setup No The paper is theoretical and focuses on mathematical derivations and bounds. It does not describe any empirical experimental setup, hyperparameters, or training configurations.