Nonlinear optimization (PhD-level), ISE 7200:
Tentative course topics:
- Convex analysis
- Preliminaries
- Convex sets, functions, and programs
- Equivalent problems
- Lagrange and Fenchel dualities
- Algorithms and their convergence analysis
- Gradient descent
- Steepest descent
- Projected gradient descent
- Accelerated gradient method
- Frank-Wolfe method
- Subgradient method
- Proximal algorithms
- Dual ascent method, Method of Multipliers (MM), Alternating Direction MM
- Primal-dual algorithms (Chambolle-Pock type)
Systems modeling and optimization for analytics (Undergraduate core course for OSU’s Data Analytics Major), ISE 3230:
Tentative course topics:
-
- Mathematical modeling for Linear Programs (LP)
- Mathematical modeling in Python (CVXPY and GurobiPy)
- Graphical solutions for LP
- Basic Feasible Solution and Extreme Points
- Algebra of the Simplex method
- Simplex in the tableau form
- Mathematical modeling for Mixed Linear Integer Programs (MILP)
- Duality for LP
- Economic interpretation of the dual (shadow prices and reduced costs)
- A few selected students’ course project videos:
-
- Rubik’s cube solver
- Sudoku solver
- Fantasy Football
- ColumSub
- Airline scheduling problem
- Optimal Cardio scheduling
- Fortnite inventory optimization
- Flight scheduling for winter break
- Traveling trick-or-treaters
- Smart portfolio optimization
- The cruise conundrum
- Santa’s stolen sleigh
- Optimizing Football schedule
- New York apartment hunt
- Optimal national park roadtrip
- Fantasy soccer
- Farming optimization
- Optimization of wildfire suppression resources
- Voter fraud
- Trail maintenance in Columbus
- Jewelry production
- Scheduling sleep & studying during COVID-19 pandemic
- Rationing groceries in quarantine
- Black Friday shopping
- Is graduation possible during COVID-19 pandemic?
- Spotify playlist selection
- Game night
- Campus food trucks
- UberEats delivery
- College student’s daily wellbeing
Machine Learning and Prescriptive Analytics (Undergraduate/Master-level), ISE 4194/6194:
Tentative course topics:
- Statistical learning
- Bias-variance tradeoff
- Brief introduction to Python
- Linear regression models
- Linear model/feature selection through regularization
- Logistic regression
- Support Vector Machines (SVM)
- Kernel methods
- Deep learning: Neural networks, Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN)
- Prescriptive analytics and decision making