Independent Work

1
Independent Work

• Spring 2008
• David Walker

2
What Is Independent Work?

• Research
• Advanced development
• Some combination of the two
• Literature survey leading to research

3
Goal Fun, Profit, Enrichment

• Opportunity to learn something in detail
• More than you could do in a standard course
• Differentiator
• Grad schools
• Jobs
• Life
• Fun
• You get to do exactly what you want to!

4
Key Points of this Talk

• Everything you need to know is on the web
• www.cs.princeton.edu/dpw/courses/iw/08-09/
• www.cs.princeton.edu/dpw/courses/iw/08-09/dates.h
  tm
• just in case Ive miscopied a date in this
  presentation, the definitive set of deadlines is
  up on the web
• But if you have questions, ask
• Donna OLeary, doleary_at_cs.princeton.edu, CS 410
• David Walker, dpw_at_cs.princeton.edu, CS 412
• And aside from logistics, the most important
  thing to do is find an advisor who you can work
  well with and a project you are excited to work
  on.

5
General Outline (1 semester)

• Find an advisor
• Start working
• 6-10 hours/week is an appropriate amount of time
  per week
• work steadily
• Give a 12 minute project proposal talk ( 3
  minutes for questions/changeover)
• Oct 6-10
• Write up a 1-page checkpoint report
• Nov 7
• 20-25 page Final Report
• Jan 5
• Poster Session
• Jan 7, 2-4pm

6
General Outline (thesis)

• Find an advisor
• Start working
• 6-10 hours/week is an appropriate amount of time
  per week
• work steadily
• Give a 12 minute project proposal talk
• Oct 6-10
• Checkpoint I
• Nov 7
• 5-7 page February Paper
• Feb 6
• Checkpoint II (Thesis Outline)
• March 27
• Final Report (40-50 pages)
• April 17
• Final Presentation
• April 20-24
• Poster Session
• May 7, 2-4pm

7
How To Have a Good 12 Weeks

• Start as soon as possible
• Take some time right now to shop around
• Read project ideas
• Read web pages
• Talk to grad students youve had as TAs
• Talk to faculty Finding an advisor you want to
  work with is the key step
• Come up with your own idea
• Be clever, but not too clever
• Do a decent sales pitch
• But listen to your advisors advice
• Dont be afraid of change
• Better to change course than dead end
• Dont change course too often

8
How To Have a Bad 12 Weeks

• Delay project selection until last minute
• Ignore your advisor
• you should try to meet once/week, even if its a
  brief meeting
• Allow yourself to get stuck
• talk to your advisor dont avoid them when you
  are stuck
• Bluff your way through checkpoints
• think hard about your plans
• work hard start early
• Give incoherent presentations
• prepare practice
• get feedback from your advisor friends
• give slides to your advisor in advance ask for
  feedback
• communication skills are the key to technical
  success

9
On to the specifics....
10
How To Find a Project

• Meet professors for lunch on Monday
• 12-130pm
• Browse the web
• wiki.cs.princeton.edu/index.php/UgradResearchTopic
  s

11
Proposal Presentations Sign up Sept 22-26
Present Oct 6-10

• Logistics
• 12 minute talk 3 minutes for questions/changeover
  
• Attend full 1-hour session and give feedback to
  other students
• Content Goals
• Describe the problem
• explain why it is important, challenging and
  interesting
• make me excited to hear more later
• Explain how the research will be evaluated
• Present a realistic plan for the semester
• point out possible stumbling points what is
  your contingency plan if research does not go as
  you hoped?

12
Proposal Presentations

• Presentation Goals
• Clearly communicate the key points
• Be sure your talk is well-structured
• Pretend you are talking to a goup of CS seniors
  who have not taken a course in the area of
  research you are pursuing
• Dont use uncommon jargon or terms without
  defining them
• Use screen real estate effectively
• Pictures, charts and graphs are your friends
• Be conscious of your presentation style
• Speak clearly
• Make eye contact
• Show energy and enthusiasm in your voice
• Assess your audiences level of comprehension
• Practice in advance and in front of friends and
  your advisor
• Heed feedback

13
Proposal Presentations

• Audience Goals
• Listen to the content and note both good and bad
  presentation elements
• both for your benefit and the speakers benefit
• Give feedback notes to your fellow students
• summarize the most important points of the talk
• what could improve
• what is already good
• For proposals, each student will attend the talks
  give in the same hour as their talk
• Grades
• Formal grades wont be given to the presentation
  itself
• However, presentation quality amount of
  progress will be a component when assigning an
  overall grade at the end of the semester.
• eg when deciding between an A- and B, well
  think about presentation quality

14
Checkpoints Nov 7/March 27

• approximately a 1 page document
• describe the accomplishments made so far
• note tasks partially or completely completed
• note papers read, experiments conducted, code
  completed
• illustrate your understanding of the topic
• give a plan for the rest of the semester
• note deadlines for completing other tasks
• meet with your advisor, have them read your
  checkpoint document and sign the checkpoint form
• You must return the form to Donna OLeary (CS
  410) by the checkpoint deadline

15
(No Transcript)
16
Poster Session Jan 7

• Create a display for a 4-by-4 bulletin board
• poster, collection of slides, etc.
• plus optional software demo
• Content
• background explain the problem motivation
• what have you learned?
• what experiments have you performed?
• what have you proven?
• what algorithms have you defined?

17
Final Report Jan 5/April 17

• Final report is 20-25 pages
• do not go much over 25 pages
• you may have an appendix that contains auxiliary
  materials such as code that spills over 25 pages
• Thesis Reports approximately twice as long
  (40-50 pages)
• Content and writing are both important
• Visuals such as graphs, diagrams, pictures, etc
  can be effective communication mechanisms
• See website for turn-in instructions specifics
  about fonts and such (which arent that
  important, but be reasonable...)
• A printed copy needs to be submitted to Donna by
  the deadline
• 2 unbound copies for a thesis
• Give your advisor a copy (either e-mail or hard
  copy as they prefer)

18
Grading

• Grades will depend upon
• the creativity and originality of student ideas
• the content, amount of work accomplished to date,
  clarity and polish of presentations
• the content, eloquence, organization and clarity
  of writing
• the majority of the grade will depend upon the
  final report
• however, poor presentations and missing
  checkpoints will also have an impact

19
Summary Most important things

• Find an adviser and project as soon as possible
• lunch on monday 12-130 may help
• feel free to email them
• check out the independent work projects page
• Go to the course web site and read the course web
  pages make a note of all deadlines
• If something is unclear, ask me, your advisor or
  Donna OLeary (doleary_at_cs.princeton.edu)
• Meet with your advisor often and talk to them as
  soon as you get stuck do not procrastinate

20
Conclusions

• A good project is a lot of work
• ... but it could be more fun than any other
  course youll take.

21
Questions?