Mechanism design augmented with output advice

Authors: George Christodoulou, Alkmini Sgouritsa, Ioannis Vlachos

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In the full version of the paper we observe the behaviour of ˆρ, η in real-world datasets [26, 7, 37, 12, 16, 3].
Researcher Affiliation Academia George Christodoulou Aristotle University of Thessaloniki and Archimedes/RC Athena, Greece gichristo@csd.auth.gr Alkmini Sgouritsa Athens University of Economics and Business and Archimedes/RC Athena, Greece alkmini@aueb.gr Ioannis Vlachos Athens University of Economics and Business and Archimedes/RC Athena, Greece ioa.vlahos@aueb.gr
Pseudocode Yes Mechanism 1 The Allocation Scaled Greedy mechanism Input: instance t Rn m, recommendation ˆa Rn m Output: a 1: rij 1 if ˆaij = 1, n β otherwise, (β [1, n]) 2: ij arg mini{rijtij} 3: if i = ij then aij = 1 else aij = 0, for each (i, j) N M
Open Source Code Yes Full code and data are provided in order to make the result reproduction possible. Data is open-source and helpful references and links are provided.
Open Datasets Yes In the full version of the paper we observe the behaviour of ˆρ, η in real-world datasets [26, 7, 37, 12, 16, 3].
Dataset Splits No The paper mentions using real-world datasets but does not explicitly specify training, validation, or test splits in the main text.
Hardware Specification Yes While there is no need for intense computational power, the details of the computing machine s CPU are included in the experimental section.
Software Dependencies No The paper does not explicitly list specific software dependencies with version numbers (e.g., Python, PyTorch, or other libraries/solvers) in its main text.
Experiment Setup No The paper discusses theoretical mechanisms and their properties but does not provide specific experimental setup details such as hyperparameter values, training configurations, or system-level settings in the main text.