Efficient and Consistent Adversarial Bipartite Matching

Authors: Rizal Fathony, Sima Behpour, Xinhua Zhang, Brian Ziebart

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

Reproducibility Variable Result LLM Response
Research Type Experimental To evaluate our approach, we apply our adversarial bipartite matching model to video tracking tasks using public benchmark datasets (Leal-Taix e et al., 2015). Our future work will explore matching problems with different loss functions and other graphical structures.
Researcher Affiliation Academia 1Department of Computer Science, University of Illinois at Chicago.
Pseudocode Yes Algorithm 1 Double Oracle Algorithm for Adversarial Bipartite Matching Equilibria.
Open Source Code No The paper does not provide an explicit statement or link for the availability of its source code.
Open Datasets Yes To evaluate our approach, we apply our adversarial bipartite matching model to video tracking tasks using public benchmark datasets (Leal-Taix e et al., 2015). Table 3. Dataset properties DATASET # ELEMENTS # EXAMPLES TUD-CAMPUS 12 70 TUD-STADTMITTE 16 178 ETH-SUNNYDAY 18 353 ETH-BAHNHOF 34 999 ETH-PEDCROSS2 30 836
Dataset Splits Yes To tune the regularization parameter (λ in adversarial matching, and C in SSVM), we perform 5-fold cross validation based on the training dataset only.
Hardware Specification No The paper does not provide specific details about the hardware used for running its experiments.
Software Dependencies No The paper mentions software like "min Conf" (Schmidt, 2008) and "SVM-Struct" (Joachims, 2008; Vedaldi, 2011) but does not provide specific version numbers for these or other software dependencies.
Experiment Setup No The paper mentions tuning regularization parameters but does not provide specific values for hyperparameters such as learning rate, batch size, number of epochs, or optimizer settings.