Transcript Formal Processor Verification

Computer Science Foundations for
Ph.D. Students
The Carnegie Mellon Perspective
Randal E. Bryant
Carnegie Mellon University
http://www.cs.cmu.edu/~bryant
CMU CS PhD Program Students
Demographics

Around 25 new students / year
 From ~800 applicants

Approximately 50% US
 Top foreign countries: India, China, Korea

Backgrounds
 Most have undergraduate or master’s degree in computer
science or related field
–2–
Program Requirements
Courses


Eight PhD-level courses
One each from list of “star” courses in following areas
 Algorithms & complexity
 Programming languages
 Artificial intelligence
 Software systems
 Computer systems
Skills


Writing, speaking, programming
Two semesters as teaching assistant
Research

–3–
Prepare & defend PhD thesis
Unusual Features of Program
No Qualifying or Comprehensive Exams

Students are admitted directly to PhD program
 Very selective admissions

Believe that courses are more useful than exams
 Exams are an unreliable measure of understanding
 Working on labs and projects more effective than reading a lot
of books and papers

Have not found qualifying exams serve intended role
 “Is student qualified to pursue a PhD?”
Student Progress Monitored Closely

Students assigned advisors after brief “Immigration” course
 Advisor serves as mentor

All students reviewed 2X/year in “Black Friday” meetings
 Student progress is collective responsibility of entire faculty
–4–
Star Courses


All assume students have
undergraduate preparation
in subject
Most courses targeted
specifically to PhD-level
students
Algorithms & Complexity


Algorithms
Complexity Theory
Artificial Intelligence



Advanced AI Concepts
Machine Learning
Planning, Execution, and
Learning
Computer Systems


Programming Languages


Type Systems for
Programming Languages
Semantics of Programming
Languages
Software Systems

Advanced Operating
Systems and Distributed
Systems

Database Management
Systems
Networking

–5–
Computer Architecture
Optimizing Compilers for
Modern Architecture
Outcomes
Graduation



Around 70% of entering students graduate
Average time between 6 & 7 years
Most students graduate as fully formed researchers
 Typically 10–20 research publications
 Ready to move right into faculty positions
Placements



Most stay in the U.S.
Academic positions
Industry research
 Microsoft Research
 IBM

Other industry
 Google
–6–
 Start-up companies
IT Workforce Issues

–7–
US IT industry is still going strong
US IT Workforce Supply
Perceptions

–8–

IT jobs are moving offshore
IT jobs are not exciting
Issues
If Undergraduate Enrollments Continue to Decline …

Supply of U.S. graduate students will be limited
 Will need to attract more non-U.S. students

Need for U.S. computer science faculty will decline
 Hiring for academic jobs will decrease
 But, will still have strong demand for PhDs from industry
–9–