SC 1 - Mathematical Programming Can Make Ecological Optimization Problems Straightforward and Tractable - Let's See How

Session Description: What is the optimal resource allocation strategy for a growing plant? For what parameters does a dynamical system best fit observed data? What interventions are best suited to protecting public health?

The answers to these questions have both scientific and social importance, but at the heart of each question lies a challenging optimization problem. Solving such problems mathematically can be daunting and approaching them with the wrong computational tools can be messy.

In this course, we'll teach a suite of tools and techniques including linear programming, convex optimization, and integer programming with the objective of helping participants quickly level up from knowing nothing about these terms to tackling sophisticated problems. Ecological applications will serve as examples used to show how these methods can be employed to incorporate necessary complexity while keeping problems solvable. We'll also discuss how to embed optimization problems in larger frameworks to, e.g., explore the consequences of altering parameters and assumptions.

Participants will derive maximum benefit from the course if they already have some experience with computer programming. Participants should bring their laptops, though we'll use browser-based notebooks for our interactive segments in lieu of having participants install programs ahead of time. All data files and code for this course will be made publicly available on github.com, so participants as well as those unable to attend will have access to them as a resource.