PhD in Complex Systems and Data Science
Nutshellfully:
Our PhD in Complex Systems and Data Science (CSDS) enables students to build a deep portfolio of research on important, complicated, data-rich problems that matter.
Students will receive a wide and rich training in empirical, computational, and theoretical methods for describing and understanding complex systems.
Depending on their chosen area of focus, students will work within and across research groups, and be strongly connected with other students through co-location and regular student-led meetings and events.
The program’s scope is science-wide, encompassing natural, artificial, and sociotechnical systems.
Program Director:
Professor Laurent Hébert-Dufresne
Application Basics
- We accept applications submitted by February 15 for Fall admission. Students may not start in the Spring.
- Students should have a relevant Master's or be able to show exceptional promise.
- We recommend prospective students identify a faculty advisor in advance.
- International students will need to apply well in advance taking into consideration visa processes.
- Please apply online through UVM's Graduate College
- Limited funding opportunities are available.
Our Essential Goal:
To train the deeply skilled and ethical research data scientists the world needs.
Our More Detailed Goal:
Our PhD is for students who want to develop a potent portfolio of scientific research into data-rich problems. We want to make this happen.
And as for our Masters, we provide students with a broad training in computational and theoretical techniques for (1) describing and understanding complex natural and sociotechnical systems, enabling them to then, as possible, (2) predict, control, manage, and create such systems.
Foundation
The PhD rests upon our Master's and our five-course Graduate Certificate.
Major skill sets we want students at all levels to develop:
- Being a good team member in a highly-collaborative pan-disciplinary, creative research environment that helps solve some of the biggest questions in the world.
- Data wrangling: Methods of data acquisition, storage, manipulation, and curation.
- Visualization techniques, with a potential for building high-quality web-based applications.
- Powerful ways of identifying and extracting explanatory, mechanistic stories underlying complex systems—not just how to use black box techniques.
Becoming the Crow
Who we're looking for to join our corvine cohort:
Students must have a Bachelor's degree and preferably a Master's degree in a relevant field and prior coursework in computer programming, calculus, linear algebra, statistics and/or probability.
Training in relevant aspects of physics (e.g., statistical mechanics) will be beneficial but not required. Applicants lacking one or more of these prerequisite areas may be accepted provisionally and will be required to complete an approved program of supplementary work within their first year of study.
Applicants will be evaluated based on their potential for excellence in research, as judged from their academic background, test scores, relevant experience and three (3) letters of recommendation. We will admit students who we believe are most likely to succeed and thrive in the program.
Required background for applicants:
Students must have prior coursework or be able to establish competency in:
- Calculus
- Coding (Python/R ideal but not necessary)
- Data structures
- Linear algebra
- Probability and Statistics
We offer three catch-up courses for students who are missing these prerequisites:
Not all three courses can be taken together:
At most one of MATH 2522 or CS 2240 may be taken for graduate credit. Students must also submit a form for pre-approval from the Graduate College at least 1 month before the semester in which they take the course.
Catch-up Course Descriptions
Applied Linear Algebra (MATH 2522)
Solving linear systems, vectors, matrices, linear independence, vector spaces, determinants, linear transformations, eigenvalues and eigenvectors, singular value decomposition, and matrix factorizations.
Data Structures (CS 2240)
Design and implementation of linear structures, trees and graphs. Examples of common algorithmic paradigms. Theoretical and empirical complexity analysis. Sorting, searching, and basic graph algorithms.
Basic Statistical Methods 1 (STAT 1410)
Foundational course for students taking further quantitative courses. Exploratory data analysis, probability distributions, estimation, hypothesis testing. Introductory regression, experimentation, contingency tables, and nonparametrics. Computer software used.
Extra pieces for admission:
GRE?
No. This is not a thing you need to take.
International Students
TOEFL score thresholds:
- Minimum for admission: 90
- Minimum to qualify for funding in a teaching assistant position at UVM: 100
You're on board? Here are the paths you can take:
You're in the PhD program so you have only one destination for your research:
Create and defend a PhD Dissertation consisting of three or more peer-reviewed journal papers. There are many ways to reach this level.
Travelers of All Paths must take the first course in each sequence in the Common Core (9 credits) and at least two accompanying second courses:
Common Core:
First Sequence
- Data Science I: CSYS/CS/STAT 5870 (required)
- Data Science 2: CSYS/CS/STAT 6870
Second sequence
- Modeling Complex Systems: CSYS/CS 6020 (required)
- Modeling Complex Systems 2: Course Number TBA
Third Sequence
- Principles of Complex Systems 1: CSYS/MATH 6701 (required)
- Principles of Complex Systems 2: CSYS/MATH 6713
(Each course scores 3 credits.)
Notch up 75 or more credits total
A minimum of seventy-five credits of graduate study must be approved by the students graduate studies committee and successfully completed. All students must take a minimum of thirty (30) credits of research and thirty (30) credits of graduate coursework, of which at least fifteen must be graded and may not count towards a Master’s degree (only courses with grades of B- or above are counted towards this minimum requirement and students with two grades below B are eligible for dismissal).
Students may transfer credits from other universities or within UVM following standard UVM policies. Students will need to earn a minimum 3.0 GPA to graduate.
Complex Systems and Data Science Electives
- Chaos, Fractals and Dynamical Systems (CSYS 5766)
- Complex Networks (CSYS/MATH 6713)
- Evolutionary Computation (CSYS/CS 6520)
- Applied Artificial Neural Networks (CSYS/CEE 7920)
- Applied Geostatistics (CSYS/STAT/CEE 7980)
- Database Systems (CS 3040)
- Human Computer Interaction (CS 3280)
- Machine Learning (CS 3540)
- Statistical Methods (STAT 3210)
- Multivariate Analysis (STAT 5230)
- Logistic Regression and Survival Analysis (STAT 5290)
- Experimental Design (STAT 5310)
- Categorical Data Analysis (STAT 5350)
- Probability Theory (STAT 5510)
- Statistical Theory (STAT 5610)
- Bayesian Statistics (STAT 6300)
- Statistical Learning (STAT/CS 3990)
Two things:
This course list evolves and not all courses will be offered in any given semester.
Other courses (including special topics) may be approved by the CSDS Curriculum Committee.
Optional elective paths to tailor your PhD
Some framing to light the possible ways:
Domain consultant: Mads Almassalkhi
- EE 5310 Electric Energy Systems Analysis
- EE 5320 Smart Grid
- CEE 3990 Reliability of Engineering Systems
- CSYS/EE 6990 Optimization in Engineering
- Other approved advanced electives related to energy
Domain consultant: Josh Bongard
- CS 3060 Evolutionary Robotics
- CSYS/CS 6520 Evolutionary Computation
- CSYS/CEE 7920 Applied Artificial Neural Networks
- BIOL 3165 Evolution
- ME 6120 Advanced Dynamics
- Other approved advanced electives related to evolutionary robotics
Domain consultant: Jason Bates
- ME 5410 Advanced Bioengineering Systems
- ME 6550 Multi-Scale Modeling
- MMG 3320 Methods in Bioinformatics
- CTS 6020 Quality in Health Care
- MATH 5788 Mathematical Biology & Ecology
- MPBP 6080 Biometrics & Applied Statistics
- Other approved advanced electives in biomedical systems related areas
Domain consultants: Donna Rizzo and Taylor Ricketts
- CSYS/STAT/CEE 7980 Applied Geostatistics
- ENVS 4990 Environmental Modeling and Systems Thinking
- Geog 3520 Advanced Topic: GIS & Remote Sensing
- Geog 3505 Spatial Analysis
- NR 4450 Integrating GIS & Statistics
- NR 6430 Fundamentals of Geographic Information Systems
- PBIO 5940 Ecological Modeling
- PBIO 5750 Global Change Ecology
- PBIO 6940 Data Modeling for Envir Science
- Other approved advanced electives related to the environment
Domain consultant: Asim Zia
- PA 6060 Policy Systems
- PA 6080 Decision Making Models
- PA 6110 Policy Analysis
- PA 6170 Systems Analysis and Strategic Management
- PA 6990 Resilient Communities: Designing at the Nexus of Food, Energy and Water Systems
- PSYS 3990 Behavioral Economics
- Other approved advanced electives related to policy
Domain consultant: Chris Skalka
- CS 3650 Computer Networks
- CS 3660 Network Security & Cryptography
- CS 3750 Mobile Apps and Wireless Devices
- Other approved advanced electives in distributed systems
Summary of the Five Stages needed to acquire Corvid Cleverness
1. Completion of 75 credits of coursework with sufficiently high enough GPA.
2. Passing of the written comprehensive exams: PhD students sit for a triplet of 1 hour written exams (in 3 subjects) on topics decided upon in consultation with their research advisor(s). For example, the exams might cover the content found in CSYS 6701, CSYS 6020, and STAT 6870. Students need not have taken all core classes prior to sitting for a particular exam, and the exams need not be taken on the same day. A follow-up oral exam could be scheduled to address material the student failed to demonstrate mastery of during the written exam.
3. Presentation and defense* of the dissertation proposal.
4. Generation of at least two published or accepted peer-reviewed publications prior to defending* their dissertation, with a third at least in peer-review. These publications must be deemed of sufficient breadth, depth, and quality by their Graduate Studies Committee.
5.Delivery of a written dissertation and oral defense* of the dissertation
*Defense Timetable for Thesis/Dissertation Students