Best Response Regression

Authors: Omer Ben-Porat, Moshe Tennenholtz

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We also test our approach in a high-dimensional setting, and show it significantly defeats classical regression algorithms in the prediction duel. The theoretical analysis is complemented by an experimental study, which illustrates the effectiveness of our approach.
Researcher Affiliation Academia Omer Ben-Porat Technion Israel Institute of Technology Haifa 32000 Israel omerbp@campus.technion.ac.il Moshe Tennenholtz Technion Israel Institute of Technology Haifa 32000 Israel moshet@ie.technion.ac.il
Pseudocode Yes Algorithm: EMPIRICAL PAYOFF MAXIMIZATION (EPM)
Open Source Code Yes Code for reproducing the experiments is available at https://github.com/omerbp/Best-Response-Regression
Open Datasets Yes The MILP algorithm is tested on the Boston housing dataset [5].
Dataset Splits Yes The dataset was split into training (80%) and test (20%) sets, and two scenarios were considered:
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments (e.g., GPU/CPU models, memory).
Software Dependencies No We employed the MILP formulation, and used Gurobi software [4] in order to find a response, where the running time of the solver was limited to one minute. The paper mentions Gurobi software but does not specify its version number.
Experiment Setup Yes We employed the MILP formulation, and used Gurobi software [4] in order to find a response, where the running time of the solver was limited to one minute.