The Trail

Document Sample
The Trail Powered By Docstoc
					                    CRA-W Graduate Student Cohort
                            March 2008

             Finding a Research Topic

                           Barbara G. Ryder
                          Rutgers University
                           Kristen Walcott
                         University of Virginia

CRA-W Grad Cohort 3/08
   •    Introductions
   •    What is CS research?
   •    How to choose an advisor?
   •    What to consider while making your choice of
   •    Focusing from area to topic
   •    How a topic is selected?
   •    Our personal experiences
   •    What to do when you are stuck?
   •    Taking risks
   •    Further discussion
CRA-W Grad Cohort 3/08
                                                       Brown U,
                         Stanford U,                   A.B. Appl Math
                         M.S. (CS) 1971                1969; wed Jon

  worked at Bell Labs,
                                                  2008, Graduates:
  Murray Hill(UNIX, C)
                                                  14 PhDs,
  1971-76; Beth 1973,
                                                  3 MS theses
  Andrew 1975

                                                  ACM Fellow
                          Career Trail            1998
Resumed Ph.D. studies
Rutgers U, 1977-82
                                           ACM SIGPLAN 1989-99
                                           (Chair, 95-97);
                                           ACM Council 2000-08
         Asst Prof,                       Full Prof,
                         Assoc Prof,
         Rutgers U                        Rutgers U
                         Rutgers U
         1982                             1994
                         Barbara - Hobbies
  • Amateur digital photographer
  • House plant lover
  • Beginning gardener
  • Classical music concert attendee
  • Reader of historical novels and/or science
  • I like to do weight training, aerobics,
    swimming, cooking, walking on the beach

CRA-W Grad Cohort 3/08
                                       About Kristen

      B.S. in CS and Math (2005)
                                                           Master of CS (2007)
      Allegheny College
      • Discovered a love of research, particularly
                                                           University of Virginia
        in software testing                                • Switched to architecture
      • Reasons for grad school:                           • Masters Project: Dynamic prediction of
             –    Get a broader view of CS                   architectural vulnerability factors
             –    Explore and solve challenging problems
             –    Desire to teach

    Hobbies                                                3rd year PhD Student
    • Dancing (ballet, jazz, modern, belly dancing)
    • Gardening
                                                           University of Virginia
                                                           • Software testing and debugging
    • Cooking
                                                           • Parallel programming challenges
    • Knitting
    • Feeding coffee and chocolate addictions
    • Convincing my friends I’m not a cat lady…

CRA-W Grad Cohort 3/08
                         What is (CS) Research?
  • Research - the systematic investigation into and
       study of materials, sources, etc., in order to
       establish facts and reach new conclusions
       (Oxford Dict.)
        – Experimental scientific research:
               •   Observe a problem
               •   Formulate a hypothesis
               •   Perform experiments and demonstrate conclusive evidence
               •   Interpret results
        – Theoretical scientific research:
               • Identification of an open question
               • Formulate a hypothesis
               • Prove hypothesis
  • Research is not knowing the answer or how to get it
CRA-W Grad Cohort 3/08
                          What is CS Research?
                         Example from Compilers
    • Code executed within loops is costly; some statements
      have same side effect on every iteration. (Observe a
    • Hypothesis: There will be performance gains if such
      code is hoisted out of loop kernals
                 • Only can do if semantic checks show no def-use links are
                   broken thereby
    • Build this optimization in a compiler and measure
      results (Gather evidence)
    • Observed gains result in invariant code motion as a
      standard compiler optimization (Interpret results)

CRA-W Grad Cohort 3/08
                         Selecting an Advisor
   • Which comes first? Advisor? Topic?
         – For many people “advisor before topic”
                • Meet faculty member with compelling research interests
         – For some people “topic before advisor”
                • Need a guide in an area already of great interest to you
   • Want an advisor
         – Knowledgeable about your topic
                • Interdisciplinary topics may require >1 advisor
         –   With compatible working style (e.g., solo vs team)
         –   With lots of research ideas
         –   With strong interest in working with PhD students
         –   ….(more this afternoon)

CRA-W Grad Cohort 3/08
                   Things to Consider in
                Selecting a Research Topic
   • Whose interest do you need to grab?
         – You
         – Your advisor
         – Your research community
   • Love your topic!
         – Sets the course for the next 2-3 years of
           your life
         – Defines area for your job search
         – May work in same/related area for years
CRA-W Grad Cohort 3/08
                    More Things to Consider
   • What are your strengths? weaknesses?
         – Programming, design, data analysis, proofs
         – Key insights vs. long/detailed verification/simulation
   • What drives you? bores you?
         – Technology, puzzles, applications, interdisciplinary
   • Do you (i.e., your advisor) have funding for you
     to work in the area?
         – Working as a TA
         – Working as an RA
         – Having university/college, government, industry, etc…

CRA-W Grad Cohort 3/08
             Focusing from Area to Topic
   • Area - usually a subfield of CS
                • Too broad to be a thesis topic
                • Determines, in part, opportunities offered to you upon
   • Topic: related open questions formulated as a
     well-defined problem in an area
         – Needs to be of compelling interest to you, and of
           interest to your advisor, and CS research community
         – Can involve both theoretical and experimental aspects
         – Best of not too narrow, to allow exploration of
           several questions, especially if some don’t work out 
         – Best if not too many other researchers on same
           approach 

CRA-W Grad Cohort 3/08
                         6 Ways to Find a Topic

CRA-W Grad Cohort 3/08
                         1) Flash of Brilliance
     • You wake up one day with a new
     • New approach to solve an important
       open problem

         • Warnings:
                – This rarely happens if at all
                – Even if it does, you may not be able to
                  find an advisor who agrees
CRA-W Grad Cohort 3/08
                         2) The Apprentice
   • Your advisor has a list of topics
   • Suggests one (or more!) that you can
     work on
   • Can save you a lot of time/anxiety
            • Warnings:
                   – Don’t work on something you find
                     boring, fruitless, badly-motivated,…
                   – Several students may be working on
                     the same/related problem
CRA-W Grad Cohort 3/08
            3) The Extended Course Project

   • You take a project course that gives you a
     new perspective
         – E.g., theory for systems and vice versa
   • The project/paper combines your
     research project with the course project
         – One (and ½) project does double duty

            • Warnings:
                  – This can distract from your research if
                    you can’t find a related project/paper
CRA-W Grad Cohort 3/08
                         4) Redo … Reinvent
   • You work on some projects
         – Re-implement or re-do
         – Identify an improvement, algorithm, proof
   • You have now discovered a topic

      • Warnings:
             – You may be without “a topic” for a long time
             – It may not be a topic worthy of a doctoral
CRA-W Grad Cohort 3/08
                          5) The Stapler
   • You work on a number of small topics that
     turn into a series of conference papers
         – E.g., you figure out how to apply a technique
           (e.g., ILP) to several key problems in an area
   • You figure out somehow how to tie it all
     together, create a chapter from each
     paper, and put a BIG staple through it
           • Warnings:
                  – May be hard/impossible to find the tie
CRA-W Grad Cohort 3/08
                         6) The Synthesis Model
   • You read some papers from other subfields in
     computer science/engineering or a related field
     (e.g., biology)
   • Look for places to apply insight from another
     (sub)field to your own
         – E.g., graph partitioning to compiler optimizations

         • Warnings:
                – You can read a lot of papers and not find
                  a connection
                – Or realize someone has done it already!
CRA-W Grad Cohort 3/08
                         … Combine, Compose…
                           but also Propose!

   • Try any combination of these ideas
         – It’s good to make sure you’re passionate
           about an area
         – BUT focus on tangible progress too
                • Are you converging to an area?
                • Have you ruled out an area?

             • Warnings:
                   – Trying these techniques can take a lot
                     of time without any results!
CRA-W Grad Cohort 3/08
                       Kristen’s experience
Undergrad:                                           Test Suite Prioritization (1st year):
• Took a seminar about how to do CS research         • Worked on conference paper 1st semester
• Became interested in software testing              • Topic was too limited and the flash of brilliance
• Undergrad thesis: Constraint-aware test suite        to extend what was started didn’t happen
  prioritization                                     • Got very bored…

Architectural Vulnerability (1st/2nd year):          Graphics (2nd year):
• Took architecture course (1st sem) and      adv.   • Had always wanted to try graphics
  architecture seminar (2 nd sem)
                                                     • Took independent study (1st sem) and adv
• Picked project for seminar using Apprentice          graphics seminar (2nd sem)
  approach                                           • Did another Extended Course Project and
• Switched advisors                                    conference paper
• Project became Extended Course Project,            • Didn’t officially switch advisors
  conference paper, and my masters project           • BUT didn’t want to have a career in graphics
• BUT didn’t really enjoy the field

Graphics + Software Testing? (2nd summer)            Testing + Parallel Programs (3rd year):
• Loved the math and prettiness of graphics and      • Heard talk on debugging parallel programs
  the applications and community in testing          • Switched to original advisor
• Met with 2 possible advisors and read a LOT        • Started another literature review
• Found a way to combine the areas                   • Now we’re trying to lay out each of the research
CRA-W no Cohort 3/08
• BUTGrad one was really excited about it              questions that I’d like to address
                         Barbara’s Experience
 • Became interested in compilers from Bell Labs
       – Took grad courses and PhD quals at Rutgers
 • Chose 1st advisor in Compiler Optimization
              • Topic - Subproblem from his thesis
              • Wanted to formulate my own research agenda
 • Changed advisors
              • Picked advisor outside my area on basis of demonstrated
                mentoring interest and research strength in algorithm design
              • Helped me formulate open research questions in SW
                evolution into a incremental dataflow analysis dissertation
              • Continuous, regular meetings throughout research period

CRA-W Grad Cohort 3/08
                Some Tips and Suggestions
   •    Topic + advisor are both important
   •    Keep an ideas “notebook”
   •    Keep an annotated bibliography listing
   •    Follow your interests and passion
         – Key driver for success and impact
                • Are you eager to get to work, continue working?
   • If not really interested, adapt
                • Is it tedium or actual lack of interest and
CRA-W Grad Cohort 3/08
                          When You’re Stuck
  • In the beginning…
        – Read/present papers regularly in a research study
          group to find open research issues
                     – Practice summarizing, synthesizing & comparing sets of papers
                     – Write your own slides for presentations
        – For a limited time, work with a senior PhD student on
          their research
        – Get feedback and ideas from others
               • Attend a top research conference in your area of interest
                     – Listen for open problems
                     – Talk to attendees about research
               • Attend your dept colloquia series and ask q’s
               • Do a government or industrial lab internship

CRA-W Grad Cohort 3/08
                         When You’re Stuck…
        – Read a PhD thesis in your area
                     – Often contain a ‘open problems’ or ‘future work’ section
        – Read your advisor’s grant proposals
        – Attend PhD oral exams and thesis defenses
                     – Understand how to formulate problems
                     – Understand what constitutes a problem solution
        – Assess your progress, with your advisor
                     – Set goals per semester - Have you ruled out an area?
                       converged on an area? Picked a topic for an exploratory
                       research project?
                     – Focus on measurable ‘good progress in an interval’ not ‘in
                       k months’ goals
CRA-W Grad Cohort 3/08
                             When You’re Stuck
   • Once started…
     – Divide your topic into milestones, and develop
       a plan to work on them one-by-one
                         – Reward yourself when you finish a milestone 
                         – Publications and/or posters as milestones
                         – Vary what you do during the day, but set aside blocks of
                           time for each activity
         – Assess your progress regularly, with your
                         – Have you submitted a workshop paper? A term project
                           with documentation? A poster at a conference? A talk at
                           a regional conf?

CRA-W Grad Cohort 3/08
                                Take Risks
  • Changing research topics
               • May move you out of your advisor’s comfort zone of
               • Have to learn the related work in a new area
               • Starting from ‘scratch’
  • Changing research advisor
               • May have to go through ‘shakedown’ period again
               • May or may not be better off
  • But change can be invigorating
               • What’s hard? Need to recognize when things are not working
                 out and take action
               • Must weigh consequences of changing and not changing

CRA-W Grad Cohort 3/08
                                   Take Risks
 • Choosing a ‘hot’ area with lots of competition in
   research community
              •   Good results ensured of impact
              •   May be easier to get funding
              •   But you may be ‘scooped’
              •   Make a context-dependent decision
 • Need to take some risk
              • Should choose significant problem
              • Reward for solution, but higher risk to obtaining solution
                    – High risk problems may not have solutions
                    – Difficult to publish negative results
 • Overall need to balance and to specialize choices
   for your situation and your interests
CRA-W Grad Cohort 3/08
            Find a topic and get started!
   Great relevant article in ACM Crossroads, “How to
     Succeed in Graduate School: A Guide for
     Students and Advisors”, (part I, Dec 1994; part
     II, Feb 1995), available in ACM Digital Library


CRA-W Grad Cohort 3/08

Shared By: