723: Mathematical Optimization, with Jerry Yurchisin
Oct 17, 2023
auto_awesome
Jerry Yurchisin, Data Science Strategist at Gurobi, discusses mathematical optimization, Gurobi solver, using Python with Gurobi, blending math optimization and ML, NLP optimization, and the application of math in sports predictions.
Mathematical optimization enhances decision making at leading enterprises like Gurobi Optimization.
Machine learning can be integrated with mathematical optimization for optimal decision support.
Linear programming applied in NLP detects plagiarism by minimizing textual differences between passages.
Deep dives
Mathematical Optimization and its Applications in Decision Making
Mathematical optimization, guided by a data science strategist, is utilized by leading companies like Gurobi Optimization for decision intelligence. They pair mathematical optimization with machine learning and statistics to inform decision making. It benefits hands-on data science practitioners, offering insights into how it works and provides real-world examples where it outperforms statistical or machine learning approaches.
Understanding Linear Programming and Mixed Integer Programming
Linear programming and mixed integer programming are explained as fundamental optimization techniques for decision making. Decision variables, constraints, and objective functions are outlined in a linear programming model. Mixed integer programming introduces binary and integer decision variables for more complex problem-solving scenarios such as NFL scheduling.
The Intersection of Machine Learning and Optimization
The relationship between machine learning and optimization is explored in three ways: leveraging machine learning for feature selection, embedding regression models within optimization frameworks, and acknowledging machine learning problems as optimization problems. This synthesis demonstrates how machine learning enhances understanding for optimal decision-making.
Unveiling Mathematical Optimization in NLP Applications
Innovative applications of mathematical optimization in NLP scenarios are exemplified through plagiarism detection using linear programming. By minimizing textual differences between passages, linear programming combined with NLP models like Google's word mover score identifies potential plagiarism instances. This application showcases the synergy between optimization and NLP in specialized use cases.
Applying Math to Sports Performance Analysis
Jerry discusses applying mathematical concepts to sports performance analysis, emphasizing the importance of understanding player output and injury likelihood in sports. He provides an example of relating physics concepts to NFL combine events to quantify athletic performance objectively. By decomposing complex data and providing new context, Jerry aims to enhance decision-making in sports analytics.
Professional Journey and Evolution into Data Science
Jerry shares his career journey from teaching math to becoming a senior mathematician and data scientist, highlighting projects in government work and sports analytics. He reflects on transitioning into consulting, showcasing projects involving mathematical optimization and statistical modeling for impactful decision-making. Through examples in cyber acquisition and asset degradation analysis, Jerry illustrates the practical applications of data science in real-world scenarios.
Mathematical optimization should be known to every data scientist: Jon Krohn speaks to Jerry Yurchisin, Data Science Strategist at Gurobi, the decision-making technology and best-kept secret of 80% of America’s leading enterprises.
This episode is brought to you by the Zerve data science dev environment, by ODSC, the Open Data Science Conference, and by CloudWolf, the Cloud Skills platform. Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.
In this episode you will learn: • What mathematical optimization is [04:27] • How Gurobi solver works [29:01] • How to use Gurobi with Python [36:08] • Coding and algebra resources [41:14] • When to use mathematical optimization and machine learning together [54:23] • Using mathematical optimization in natural language processing [1:01:00]
