Siyuan Sun

sysun [at] berkeley [dot] edu

sysun [at] berkeley [dot] edu

TuTh 2-330P, 56 Barrows

Course on optimization; course on statistics or stochastic processes

4 homeworks (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 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 (one page summary) is due on March 3, 2015**. The **project report (10-12 pages) is due on the last day of lecture May 7, 2015**. 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 20
- Ordinary Least Squares; Course Syllabus
- Jan 22
- Local Linear Regression
- Jan 27
- Cross Validation
- Jan 29
- M-Estimators
- Feb 03
- Abstract Structure
- Feb 05
- Lasso Regression
- Feb 10
- Semiparametric Models
- Feb 12
- Nadaraya-Watson
- Feb 17
- Other Models
- Feb 19
- Stability
- Feb 24
- Observability
- Feb 26
- Parametric Optimization
- Mar 03
- Reachability
- Mar 05
- Linear MPC
- Mar 10
- Tube MPC
- Mar 12
- Learning-Based MPC
- Mar 17
- The Oracle
- Mar 19
- Epi-Convergence
- Mar 31
- L2NW in LBMPC
- Apr 02
- Semiparametric Regression for HVAC; Hybrid System Models of HVAC
- Apr 07
- LBMPC Control of HVAC; Comparing HVAC Controllers
- Apr 09
- Extensions of LBMPC for Quadrotors

- Mar 17
- Homework 1 – Due Thursday, April 2, 2015

winequality-red.csv - Apr 07
- Homework 2 – Due Tuesday, April 14, 2015
- Apr 21
- Homework 3 – Due Tuesday, April 28, 2015