MWF 12-1P, in 3102 Etcheverry

Course on optimization; course on statistics or stochastic processes

(About 3-4) programming homeworks in R (50%); class project (50%)

This course will cover topics related to the interplay between optimization and statistical learning. The first part of the course will cover statistical modeling procedures that can be defined as the minimizer of a suitable optimization problem. The second part of the course will discuss the formulation and numerical implementation of learning-based model predictive control LBMPC, which is a new method for robust adaptive optimization that can use machine learning to provide the adaptation. The last part of the course will deal with inverse decision-making problems, which are problems where an agent's decisions are observed and used to infer properties about the agent.

The projects can be in the form of a literature review, a comprehensive application of data analysis methods, or involve the exploration of original research ideas. The project should be chosen in consultation with the course instructor, and a **project proposal (two page summary) is due on March 5, 2014**. The **project report (10-12 pages) is due on the last day of classes May 2, 2014**. Joint projects, involving reasonably sized groups, are allowed.

Specific topics that will be covered include:

- Regression – Classical M-estimators; high-dimensional M-estimators; collinearity; semiparametric regression and Nadaraya-Watson regression
- Learning-Based Model Predictive Control (LBMPC) – Robustness; consistent approximations; oracle design; software code generation
- Inverse Decision-Making Problems – Inverse reinforcement learning; learning objective/utility functions; Learning utilities from game-theoretic equilibria described by variational inequalities

- Jan 22
- Ordinary Least Squares; Course Syllabus
- Jan 24
- Bias-Variance Tradeoff
- Jan 27
- Stochastic Convergence
- Jan 29
- Cross-Validation
- Jan 31
- Classical M-Estimators
- Feb 3
- Collinearity
- Feb 5
- Abstract Structure
- Feb 7
- Lasso Regression
- Feb 10
- Extensions of Lasso
- Feb 12
- Nadaraya-Watson
- Feb 14
- Semiparametric Models
- Feb 19
- Partially Linear Models
- Feb 21
- Other Models
- Feb 24
- Stability
- Feb 26
- Controllability
- Feb 28
- Observability
- Mar 3
- Parametric Optimization
- Mar 5
- Lyapunov Theory
- Mar 7
- Reachability
- Mar 10
- Linear MPC
- Mar 12
- Robust Reachability
- Mar 14
- Tube MPC
- Mar 17
- Robustness in Tube MPC
- Mar 19
- Learning-Based MPC
- Mar 21
- The Oracle
- Mar 31
- Selected Variational Analysis
- Apr 2
- Consistent Approximation
- Apr 4
- L2NW in LBMPC
- Apr 7
- Semiparametric Regression for HVAC
- Apr 9
- Hybird System Models of HVAC
- Apr 11
- LBMPC Control of HVAC; Comparing HVAC Controllers
- Apr 14
- Extensions of LBMPC for Quadrotors
- Apr 16
- Extensions of LBMPC for Quadrotors; Experiments of LBMPC on Quadrotor
- Apr 18
- Experiments of LBMPC on Quadrotor
- Apr 21
- Numerical Implementation of Linear LBMPC
- Apr 23
- Inverse Decision Making
- Apr 25
- Estimating an Individual Utility
- Apr 28
- Details for Single Utility Learning
- Apr 30
- Estimating Multiple Utilities
- May 2
- Details for Multi-Agent Utility Learning

- Mar 3
- Homework 1 – Due Monday, March 31, 2014

winequality-red.csv - Apr 14
- Homework 2 – Due Monday, April 21, 2014
- Apr 21
- Homework 3 – Due Monday, April 28, 2014