English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In deterministic optimization, it is typically assumed that all problemparameters are fixed and known. In practice, however, some parameters may be apriori unknown but can be estimated from historical data. A typicalpredict-then-optimize approach separates predictions and optimization into twostages. Recently, end-to-end predict-then-optimize has become an attractivealternative. In this work, we present the PyEPO package, a PyTorchbasedend-to-end predict-then-optimize library in Python. To the best of ourknowledge, PyEPO (pronounced like pineapple with a silent "n") is the firstsuch generic tool for linear and integer programming with predicted objectivefunction coefficients. It provides four base algorithms: a convex surrogateloss function from the seminal work of Elmachtoub and Grigas [16], adifferentiable black-box solver approach of Pogancic et al. [35], and twodifferentiable perturbation-based methods from Berthet et al. [6]. PyEPOprovides a simple interface for the definition of new optimization problems,the implementation of state-of-the-art predict-then-optimize trainingalgorithms, the use of custom neural network architectures, and the comparisonof end-to-end approaches with the two-stage approach. PyEPO enables us toconduct a comprehensive set of experiments comparing a number of end-to-end andtwo-stage approaches along axes such as prediction accuracy, decision quality,and running time on problems such as Shortest Path, Multiple Knapsack, and theTraveling Salesperson Problem. We discuss some empirical insights from theseexperiments, which could guide future research. PyEPO and its documentation areavailable at https://github.com/khalil-research/PyEPO.

PyEPO: A PyTorch-based End-to-End Predict-then-Optimize Library for Linear and Integer Programming