Towards Minimax Online Learning with Unknown Time Horizon

Authors: Haipeng Luo, Robert Schapire

ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments show that our algorithm outperforms many other existing algorithms in an online linear optimization setting.
Researcher Affiliation Academia Haipeng Luo HAIPENGL@CS.PRINCETON.EDU Department of Computer Science, Princeton University, Princeton, NJ 08540 Robert E. Schapire SCHAPIRE@CS.PRINCETON.EDU Department of Computer Science, Princeton University, Princeton, NJ 08540
Pseudocode No The paper describes algorithms verbally and mathematically, but it does not include a clearly labeled 'Pseudocode' or 'Algorithm' block/figure.
Open Source Code No The paper does not provide any explicit statements about releasing source code or links to a code repository for the methodology described.
Open Datasets No The paper describes experiments conducted in a simulated online linear optimization setting rather than using a named, publicly available dataset. It states, 'we conduct an experiment that compares the regrets of four algorithms at any time step within 1000 rounds against an adversary that chooses points in S uniformly at random (N = 10).'
Dataset Splits No The paper describes an online learning experiment setup where data is generated sequentially, rather than using traditional train/validation/test splits on a static dataset. No specific dataset split percentages or sample counts for validation are provided.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory, cloud resources) used to run the experiments.
Software Dependencies No The paper describes the algorithms and their theoretical properties but does not list specific software libraries or their version numbers used for implementation or experimentation.
Experiment Setup Yes In Section 6.1, the paper states: 'we conduct an experiment that compares the regrets of four algorithms at any time step within 1000 rounds against an adversary that chooses points in S uniformly at random (N = 10).' It also describes specific algorithmic parameters, such as 'OGD, with parameter = 2/t' and for the proposed algorithm 'continuous random variable with probability density f(T) / 1/T^2'.