Learning Algorithms for Second-Price Auctions with Reserve

Authors: Mehryar Mohri, Andres Munoz Medina

JMLR 2016 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present novel algorithms for solving this problem and report the results of several experiments in both synthetic and real-world data demonstrating their effectiveness.
Researcher Affiliation Academia Mehryar Mohri and Andr es Mu noz Medina Courant Institute of Mathematical Sciences 251 Mercer Street New York, NY, 10012
Pseudocode Yes Algorithm 1 Sorting
Open Source Code No The paper discusses the methodology and presents experimental results but does not provide any explicit links to source code or statements about its public release for the work described.
Open Datasets Yes We were able to procure an e Bay data set consisting of approximately 70,000 secondprice auctions of collector sport cards. The full data set can be accessed using the following URL: http: //cims.nyu.edu/ munoz/data.
Dataset Splits Yes We randomly sampled 2,000 examples for training, 2,000 examples for validation and 2,000 examples for testing.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper mentions the use of 'off-the-shelf QP solvers' for quadratic-programming but does not specify any particular software or version numbers.
Experiment Setup Yes For all our experiments, the parameters λ, γ and α were selected respectively from the sets {2i|i [ 5, 5]}, {0.1, 0.01, 0.001}, and {0.1, 0.2, . . . , 0.9} via validation over a set consisting of the same number of examples as the training set. Our algorithm was initialized using the best solution of the convex surrogate optimization problem.