Still under construction
This quals guide is for people with the main focus of "Signals" and "Systems". It is well known that these areas are traditionally harder than the other popular areas. However, with adequate preparation, there should be no problems passing quals with these as your primary areas.
Thanks to my quals buddies for many of the ideas on this page and their support during quals! Jaime, Derek, Frank, Qiuhua, Shangping, Yu, Yifan, Wendy, Eddy and Mandy.
Unless you are a born genius, the single most important preparation task is to form or join a quals group. The purpose of a quals group is to practice answering questions under pressure and also hopefully have a supportive environment for when the pressure of quals hits.
How to form a group: The most common way is to leave your contact details on the quals forums. Another good way is to try and find older students to recommend a group. Most people join multiple quals groups in order to get more practice and motivation. This also allows you to leave one if you don't click with the other members.
Ideal size: The idea size of a group is 4 or 6 people. Odd numbered groups will have trouble running quals simulations. Groups with many people will be bogged down with long meetings.
How often should meetings be held?: At least one or two meetings per week staring early in autumn quarter is a minimum. The frequency usually picks up after thanksgiving. Each meeting lasts about 2 hours. Any longer and people start falling asleep.
Meeting format: A two phase format is common and seems to works well. Starting with the older quals questions or quals like questions obtained from elsewhere, each person prepares a solution to different question. In phase one, each person goes to the whiteboard and presents the question, 10-12 minutes is given to solve the problem and then the solution is given. Then 10 min is spent discussing the solution and the key concepts to remember if required.
After thanksgiving phase two can begin. This is the simulation phase. Again each person prepares the solution to a quals problem at home. Members go off in pairs to corners of the room and take turns in acting as the professor and the student. 10-12 min is allowed per question and there is no discussion until everyone has answered every question. In this stage it is helpful to consult with a student who has taken quals on how each professor acts during the exam.
I'll add this when I have time, formula sheets, useful books, etc…
Signals and Systems Summary [incomplete]
Choosing professors is overly complicated. Essentially you select 4 groups of 5 professors, this goes into Prof. Gill's magical algorithm and you get your final 10 examiners. Despite this apparent black box there are some strategies that can be used to optimize your selection.
How does it work?: There are 4 groups labeled 1,2,3 and 4, where 1 is your first preference and 4 is your last preference. In general, four professors will be selected from group 1, three from group 2, two from group 3 and 1 from group 4. Hence, you ideally put your most favored professors in group 1 and so on. A further restriction is that only 6 professors from each area can be chosen.
Main strategy: The main strategy in selecting professors is to exploit the fact that every professor is human and has a limited amount of students he or she can examine in one week. Some professors are well known to be popular and are selected in group 1 and 2 by many students (You can ask older students who they are). As a result, their quota is quickly exhausted. Hence, if a popular professor is selected in group 3 or 4, it is highly unlikely that you will get that professor as an examiner. You can use this to your advantage by selecting some popular professors not in your area of expertise in groups 3 and 4 since you will not get them anyway. This will boost the probability of getting the other less popular professors you selected in those groups.
Example confirmation email:
Quals preference form for Submitted at 2011-01-04 20:42:40 Group 1: Thomas Cover, Antony Fraser-Smith, Hector Garcia-Molina, Tsachy Weissman, Howard Zebker Group 2: John Gill, Andrea Goldsmith, Robert M. Gray, Dwight Nishimura, Balaji Prabhakar Group 3: Donald Cox, Brad Osgood, John Pauly, Julius O. Smith, Benjamin Van Roy Group 4: Stephen Harris, Lambertus Hesselink, Philip Levis, Nick McKeown, Fouad Tobagi
About 3 days before your exam you will receive the timetable for your examination day.
Example censored timetable email: Everyone gets this email at the same time. This happens when the examination schedule has been finalized.
Your qualifying examinations are scheduled for Friday, January 14, 2011. The times of your ten examinations are shown below. 1 Friday 10:00-10:12 2 Friday 10:24-10:36 3 Friday 10:48-11:00 4 Friday 11:12-11:24 5 Friday 11:36-11:48 6 Friday 1:36- 1:48 7 Friday 2:12- 2:24 8 Friday 2:36- 2:48 9 Friday 3:12- 3:24 10 Friday 3:48- 4:00 The names of the examiners will be sent by email shortly after noon on Tuesday, January 11. The number of candidates scheduled this year is 154. If you want to withdraw from the quals, you may do so without penalty any time before your first exam.
Example final timetable: You get this about 3 days before your examination day. This date is different for everyone.
The Ph.D. Qualifying Examination consists of ten 12-minute sessions, all on the same day. Your schedule is shown below. It is recommended that you visit your examiners' offices before your exam day so that you will not have trouble locating offices on exam day. Do not discuss the contents of the examinations with anyone until the examination week is over. Do not ask your examiners for an evaluation of your performance. Reminder: examiners may ask questions from their specified question areas or from fundamental areas of mathematics, physics, or common sense. Candidate: SUID: Friday, January 14, 2011 # Day Time Examiner Office Area Group 1 Friday 10:00-10:12 John Gill Packard 266 Sig 2 2 Friday 10:24-10:36 Balaji Prabhakar Packard 269 Sig 2 3 Friday 10:48-11:00 Hector Garcia-Molina Gates 434 S/W 1 4 Friday 11:12-11:24 Robert M. Gray Packard 261 Sys 2 5 Friday 11:36-11:48 Donald Cox Packard 361 E&M 3 6 Friday 1:36- 1:48 Thomas Cover Packard 254 Sig 1 7 Friday 2:12- 2:24 Antony Fraser-Smith Packard 358 E&M 1 8 Friday 2:36- 2:48 Philip Levis Gates 358 S/W 4 9 Friday 3:12- 3:24 Benjamin Van Roy Huang 356 Sig 3 10 Friday 3:48- 4:00 Howard Zebker Packard 334 Sys 1
How to prepare for examination day: Just treat it like any normal exam. Do your final reviews and get lots of sleep.
What to wear on the day: Just wear something neat and comfortable. You don't need to go overboard and wear a suit jacket. Just remember that you will need to do a lot of walking!
What to expect from the examiners: This really depends on the specific examiner. You should try to find out from past students what how your examiners will examine you. Some will be very formal and make you answer a list of questions. Others will simply have a chat with you. I don't think it is appropriate to post profiles of examiners that I have collected over the years here so you will have to try and find out from your friends.
Other questions?: Email me! I will post the answers here if appropriate.
There is about a 2 week wait before the results are released. Usually they are released in the afternoon (about 5pm) after the committee meeting.
What happens if I pass?: Congratulations! Just follow the instructions and you are fine.
What happens if I don't pass?:For people who are taking it the first time, this will generally mean you will have to take it again. Only in very rare cases will your advisor file a petition if you don't pass the first time. If it is the second time, then generally your advisor will file a petition for you. So don't worry!
Example results email:
Congratulations! By action of the faculty of the Electrical Engineering Department, you have passed the Ph.D. Qualifying Examination. Your next step is to apply for candidacy for the Ph.D degree. To complete the application for candidacy, you must receive formal sponsorship of two faculty members, as thesis supervisor and associate supervisor, and file an acceptable program of study. The application should be submitted by the end of spring quarter of the next academic year, 2011-2012. Enclosed are your examination scores and your examiners' statistics. Name: SUID: Score: 71.2 Rank: 27 / 147 Examiner Adj Raw 25% 50% 75% Grp Area ------------------------------------------------------- Thomas Cover 8.7 8.5 4.5 5.8 8.0 1 Sig Howard Zebker 7.5 7.5 4.0 7.0 8.0 1 Sys Antony Fraser-Smith 6.7 7.0 5.2 7.0 8.0 1 E&M Hector Garcia-Molina 5.9 5.0 3.2 5.0 7.0 1 S/W Robert M. Gray 8.1 8.0 2.0 5.0 8.2 2 Sys Balaji Prabhakar 7.5 7.2 4.5 7.0 8.8 2 Sig John Gill 6.1 6.0 4.9 6.8 7.0 2 Sig Benjamin Van Roy 8.2 8.0 5.0 6.5 8.0 3 Sig Donald Cox 7.4 7.5 5.0 6.0 8.0 3 E&M Philip Levis 5.1 5.0 4.5 6.0 8.2 4 S/W Quals statistics: Median 61.8. Mean 60.5. Std Dev: 12.7. Passing rank 84. Passing score 58.9. Explanation of Scores Adj Normalized score after adjusting examiner's mean and variance Raw Raw score reported by examiner 25% 25% of raw scores for this examiner were lower than this score 50% Median raw score for this examiner 75% 75% of raw scores for this examiner were lower than this score Group Candidate's ranking of this examiner; group 1 is top ranked group