Adaptive Online Learning in Dynamic Environments

Authors: Lijun Zhang, Shiyin Lu, Zhi-Hua Zhou

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

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we study online convex optimization in dynamic environments, and aim to bound the dynamic regret with respect to any sequence of comparators. ... we develop a novel online method, namely adaptive learning for dynamic environment (Ader), which achieves an optimal O( p T(1 + PT )) dynamic regret. The basic idea is to maintain a set of experts, each attaining an optimal dynamic regret for a specific path-length, and combines them with an expert-tracking algorithm. ... We establish the first lower bound for the general regret bound in (2), which is Ω( p T(1 + PT )). We develop a serial of novel methods for minimizing the general dynamic regret, and prove an optimal O( p T(1 + PT )) upper bound.
Researcher Affiliation Academia Lijun Zhang, Shiyin Lu, Zhi-Hua Zhou National Key Laboratory for Novel Software Technology Nanjing University, Nanjing 210023, China {zhanglj, lusy, zhouzh}@lamda.nju.edu.cn
Pseudocode Yes Algorithm 1 Ader: Meta-algorithm; Algorithm 2 Ader: Expert-algorithm; Algorithm 3 Improved Ader: Meta-algorithm; Algorithm 4 Improved Ader: Expert-algorithm; Algorithm 5 Ader: Expert-algorithm with dynamical models
Open Source Code No The paper does not provide any specific statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper does not describe any experiments involving datasets, training, or data access information.
Dataset Splits No The paper does not describe any experimental setup involving dataset splits for training, validation, or testing.
Hardware Specification No The paper focuses on theoretical contributions and does not mention any specific hardware specifications used for experiments.
Software Dependencies No The paper is theoretical and does not describe any software dependencies with specific version numbers.
Experiment Setup No The paper does not detail any experimental setup, hyperparameters, or system-level training settings as it focuses on theoretical development.