Improved Decision Making with Pareto Fronts

Fri September 10, 04:15 PM–04:45 PM • Back to program
Start time 16:15
End time 16:45
Countdown link Open timer

Optimising for multiple objectives is a non-trivial task, especially when they are in conflict. For example how can one best overcome the classic trade-off between quality and cost of production, when the monetary value? In this gentle introduction to multi-objective optimisation you will learn about Pareto Fronts and how they can be used to optimise for multiple objectives.

As a data driven decision maker you will learn the advantages and shortcomings of the technique and be able to assess applicability for your own projects.

Optimising for multiple objectives is a non-trivial task, especially when they are in conflict. For example, how can one best overcome the classic trade-off between quality and cost of production, when the monetary value of quality is not defined? In this talk you will learn about Pareto Fronts and how they can be used to optimise for multiple objectives simultaneously.

When applicable, this Pareto Optimisation method provides better results than the common practice of combining multiple parameters into a single parameter heuristic. The reason for this is quite simple. The single heuristic approach is like horse binders limiting the view of the solution space, whereas Pareto Optimisation enables a bird’s eye view.

Real world applications span from supply chain management, manufacturing, aircraft design to land use planning and therapeutics discovery.

This introduction is geared towards anyone who makes data driven decisions or is interested in improving their optimisation skills. You will learn the advantages and shortcomings of the technique and be able to assess applicability for your own projects.

Specifically I will demonstrate that the Python package DEAP is a useful tool for prototyping for solutions in large intractable spaces, such as DNA combinatorial spaces, which require stochastic search techniques such as Genetic Algorithms.

Eyal Kazin

Ex-Cosmologist turned Data Scientist with over 10 years experience in machine learning, statistical inference, and passionate about data insights visualisations. Result driven and highly detail oriented I’m motivated by intellectual challenges and love analysing data to communicate insights for better decisions within organisations.

My claim for fame is paying rent in four different continents within a span of a decade, including three tennis Grand Slam cities (NY, Melbourne, London).