Towards Utilitarian Combinatorial Assignment with Deep Neural Networks and Heuristic Algorithms

Dividing the Indivisible : Algorithms, Empirical Advances, and Complexity Results for Value-Maximizing Combinatorial Assignment Problems

“…The distributions that we use for benchmarking throughout this section are NPD (normal probability distribution), UPD (uniform probability distribution), and NDCS (normally distributed coalition structures, drawing inspiration from [152]). These generate some of the most difficult known problems for the state-of-the-art UCA solvers (see e.g., our papers [143,144] for more examples). Variations of them have also been used extensively to benchmark various approaches for solving a number of different problems related to UCA in operations research, algorithmic game theory, and multi-agent systems, including coalition formation and combinatorial auctions [121,146,144,142,152,153,171,218].…”

“…A batch size of 32 was used for the neural networks, together with 100 epochs, and finding their hyperparameters took us 4 days, running 44 experiments concurrently. Our hyperparameter search space together with the final hyperparameters are presented in [143].…”