Robust Budget Allocation via Continuous Submodular Functions

Authors: Matthew Staib, Stefanie Jegelka

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our Robust Budget Allocation algorithm on both synthetic test data and a real-world bidding dataset from Yahoo! Webscope (yah) to demonstrate that our method yields real improvements. For all experiments, we used Algorithm 1 as the outer loop. For the inner submodular minimization step, we implemented the pairwise Frank-Wolfe algorithm of (Lacoste-Julien & Jaggi, 2015).
Researcher Affiliation Academia Matthew Staib 1 Stefanie Jegelka 1 1Massachusetts Institute of Technology.
Pseudocode Yes Algorithm 1 Subgradient Ascent
Open Source Code Yes Our code is available at git.io/v HXk O.
Open Datasets Yes To evaluate our method on real-world data, we formulate a Budget Allocation instance on advertiser bidding data from Yahoo! Webscope (yah). This dataset logs bids on 1000 different phrases by advertising accounts. ... Yahoo! Webscope dataset ydata-ysm-advertiser-bids-v1 0. URL http://research.yahoo.com/ Academic_Relations.
Dataset Splits No The paper evaluates on 'synthetic test data' and 'real-world bidding dataset from Yahoo! Webscope', but it does not specify train/validation/test dataset splits, percentages, or absolute sample counts for these splits.
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. It only vaguely mentions 'MIT Supercloud and the Lincoln Laboratory Supercomputing Center'.
Software Dependencies Yes implementing the linear oracle using MOSEK (MOSEK Ap S, 2015). (In references): MOSEK Ap S. MOSEK MATLAB Toolbox 8.0.0.57, 2015. URL http://docs.mosek.com/8.0/ toolbox/index.html.
Experiment Setup Yes For all experiments, we used Algorithm 1 as the outer loop. For the inner submodular minimization step, we implemented the pairwise Frank-Wolfe algorithm of (Lacoste-Julien & Jaggi, 2015). In all cases, the feasible set of budgets Y is {y 2 RS s2S y(s) C} where the specific budget C depends on the experiment. Our code is available at git.io/v HXk O. For both settings, we set |S| = 6 and |T| = 2 and discretize with δ = 0.001.