Efficiently Approximating the Pareto Frontier: Hydropower Dam Placement in the Amazon Basin

Authors: Xiaojian Wu, Jonathan Gomes-Selman, Qinru Shi, Yexiang Xue, Roosevelt Garcia-Villacorta, Elizabeth Anderson, Suresh Sethi, Scott Steinschneider, Alexander Flecker, Carla Gomes

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

Reproducibility Variable Result LLM Response
Research Type Experimental We provide empirical results showing that our methods outperform other approaches in efficiency and accuracy. Our work is motivated by a problem in computational sustainability concerning the proliferation of hydropower dams throughout the Amazon basin.Table 1: The equivalent DP binary tree (Western Amazon) leads to considerable speed-ups:Table 2: Examples of runtimes and number of solutions:Figure 5: Approximation quality:Figure 6: Compared to the MIP approximations:
Researcher Affiliation Academia Xiaojian Wu Dept of Computer Science Cornell University xw458@cornell.edu Jonathan Gomes-Selman Dept of Computer Science Stanford University jgs8@stanford.edu Qinru Shi Dept of Computer Science Cornell University qs63@cornell.edu Yexiang Xue Dept of Computer Science Cornell University yexiang@cs.cornell.edu Roosevelt Garc ıa-Villacorta Dept of Ecology and Evolutionary Biology Cornell University rg676@cornell.edu Elizabeth Anderson Dept of Earth & Environment Florida International University epanders@fiu.edu Suresh Sethi U.S. Geological Survey New York Cooperative Fish and Wildlife Unit Cornell University suresh.sethi@cornell.edu Scott Steinschneider Dept of Biological & Environ. Engr. Cornell University ss3378@cornell.edu Alexander Flecker Dept of Ecology & Evolutionary Biology Cornell University asf3@cornell.edu Carla P. Gomes Dept of Computer Science Cornell University gomes@cs.cornell.edu
Pseudocode No The paper describes algorithms (DP, MIP) and mathematical formulations but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper does not explicitly state that the source code for the methodology described is publicly available, nor does it provide a link to a code repository.
Open Datasets Yes Data: In order to test our methodological approach at different spatial scales, we developed three sub-sets of data that correspond to different geographical areas within the Amazon region: the western Amazon basin, the Mara non basin, and the entire Amazon basin (Finer and Jenkins 2012; Shedlock et al. 2000; Venticinque et al. 2016; Winemiller et al. 2016; Zarflet al. 2015). We use the Amazon river network for the tree-structured graph and corresponding values for connectivity, sediment, and seismic risk, as described in section 2.
Dataset Splits No The paper does not explicitly provide training/validation/test dataset splits. It evaluates its methods on different geographical basins as separate instances rather than splits of a single dataset.
Hardware Specification No The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments.
Software Dependencies No The paper mentions 'MIP solvers like CPLEX' and 'Non-dominated Sorting Genetic Algorithm (NSGA-II)' but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes The paper specifies values for the approximation parameter ϵ, such as 'ϵ = 0.1', 'ϵ = 0.2', and 'ϵ = 0.25' when describing the performance of the DP and MIP approaches. These are concrete parameters for their algorithms.