Pervasive Parallelism
Course Descriptor for formal details.
Overview
The Pervasive Parallelism (PP) course has three aims:
- To help you develop your MSc research proposal. The research proposal document
is the only piece of work which will be formally assessed for this course.
- To help you to develop research skills in reading and condensing research literature,
formulation of research proposals and in presentation of research.
- To expose you to the breadth of research issues involved in the pervasive parallelism
challenge, beyond your own intended specialist area.
In addition to your normal, regular meetings with your supervisors, PP will require you to work
together with other members of the class, to prepare and deliver two short presentations. The format of these
interactions will be novel: you will research and prepare presentations in your own specialist
area, but you will also deliver a presentation prepared by another member of the class.
More specifically, for the first presentation, you will deliver the material prepared by one of
your classmates, and your material will be delivered by another, different, classmate. The pairings will
be changed for the second presentation, so that overall, you will work with four other members of the
class. For the second presentation you will deliver your own material, but will work closely with a classmate during
its development.
Timetable
Tuesday meetings are in room 4.1, Lister Learning and Teaching Centre, Roxburgh Street, and Friday meetings are in room G.01, 50 George Square.
(see
Campus Map).
The First Presentation
In discussion with your supervisor, you should select two research
papers in your area of specialist interest. You should prepare a 10 minute presentation on these
(that's 10 minutes in total, not 10 for each paper), plus 2 minutes for questions. While
preparing the presentation, you should meet your 'presenter' (i.e. the person who has been
allocated to present your work), to explain and discuss the papers and the presentation you have
prepared. Please meet your presenter as early in the process as possible, even before you are sure
which papers you will choose. At the same time, you will be preparing to act as presenter for
some other classmate.
The presentation should cover
- Context. How does this work fit in to the bigger research landscape?
- Key Contribution. How do the papers advance knowledge in the area? What
are the key insights, clever ideas, unusual discoveries, as appropriate to the
content?
- Impact. Have the papers received much attention? Where were they originally
published? Where and how have they been cited or otherwise noted?
- Critical Reflection. What are you own thoughts on the work, positive and
negative, and how does it relate to what you may eventually want to research?
After each presentation session, please e-mail your comments on the presentations you have
seen to Murray Cole (mic@inf.ed.ac.uk). These can include comments on the material itself
and on the format and style of the presentation. I will filter, summarise and anonymise the
comments and pass them back to the presenter and the originator of the material. Please
make sure that your comments are constructive and helpful. The intention is that you help
your fellow students to reflect upon and improve their research and presentation skills, in a
mutually supportive environment. As noted previously, the presentations are not
formally assessed.
The Second Presentation
The delivery arrangements are similar to those for the
first presentation, i.e. 10 minutes in total, plus 2 minutes for questions, but the material
and partnering is different. This time you should prepare a presentation on your own MSc research proposal, as it evolves, which you will present yourself. However, you should work closely with your assistant, including giving at least one dry-run presentation, as you prepare the material.
The presentation should cover
- Purpose. What is the problem to be addressed? Why is this important
to fellow researchers, industry or society?
- Background. How does previous related work address (or fail to address)
this problem?
- Methods. What will you actually do? How will this shed light on the
problem?
- Evaluation. How will you decide if you have been successful? What will you
measure, prove or demonstrate?
Comments and feedback will be gathered and circulated as for the first presentation.
The presentation will take place
before the deadline for submission of the proposal. This is to allow you to reflect upon
the feedback to your presentations, and improve your proposal as a result.
Your Research Proposal
Your Research Proposal is the plan for your MSc project. The better the plan, the easier the
project is to carry out. As you develop the plan, reflect on it and evaluate it against the marking
criteria discussed below. The more critical you are now, the easier it will be to get the work done.
The research proposal should be about 6 pages long. A good proposal will provide a convincing case
for the high quality of the proposed research. It will show an awareness of relevant prior
work and include a clear statement of the problems and hypotheses to be addressed and why
they are important. It must also make clear exactly how the methods used to research
those hypotheses will yield interesting results. There are many ways in which one
might structure the material. As a guide, a good proposal might be organised as follows:
- Purpose: a statement of the problem to be addressed. This should include arguments
as to why solving the problem is important; e.g., because it will enable certain
applications, or lead to interesting scientific discoveries.
- Background: a short description of how previous work addresses (or fails to address)
this problem, leading to a rationale for the hypotheses that you intend to test,
and a convincing argument about how that hypotheses might solve the problem.
- Methods: A description of the methods and techniques you intend to use to
test your hypotheses (e.g., data analysis procedures, experimental design etc),
indicating that alternatives have been considered and ruled out on sound scientific grounds.
- Evaluation: Details of the metrics by which you will evaluate the outcomes of your
research; e.g., by comparing the output of your system with some gold standard,
or with the ways in which humans perform a task, etc.
- Outputs: A description of what the outputs of the projects will be: e.g., these might
include an extension or change to some existing theory or to some piece of software,
some new data and so on.
- Workplan: A timetable or research plan, detailing what will be done to complete
the proposed project, and when these tasks will be completed by.
Submission
Please submit your proposal as a pdf file, using the DICE command
submit perp 1 filename.pdf
where filename.pdf is the name of your file.
Good Scholarly Practice. Please remember the University requirement as regards all assessed work. Details about this can be found under academic misconduct. Furthermore, you are required to take reasonable measures to protect your assessed work from unauthorised access. For example, if you put any such work on a public repository then you must set access permissions appropriately (generally permitting access only to yourself, or your group in the case of group practicals).
Late Submission. Please be aware of the School policy on late coursework submission.
Assessment and Feedback
You will receive two forms of feedback during the course:
- Comments gathered from the presentation audience, filtered and summarised
by the course organiser, and returned to you by e-mail.
- Written feedback and a mark for your research proposal, from your supervision
team. University regulations dictate that the mark itself will not be available until
after the exam board (i.e. as though it were an exam mark), but formative feedback will
be available by the deadline in the timetable above.
Your research proposal will be marked by your supervisor in discussion with your co-supervisors.
It will be awarded a mark against the University Common Marking Scheme (CMS4). In coming to a mark, the markers will consider a number of criteria, as follows:
- Clear problem or hypothesis: the proposed topic is well-defined in the sense that there is a clear hypothesis to be tested or a particular component to be built with clear characteristics or some other characterisation of what is to be done.
- Topic relevant to Pervasive Parallelism: it should be clear that the topic is of interest to
the field of Pervasive Parallelism and is not an application that will only have interest in some other field for which realising it only requires routine use of computing.
- Motivated and significant: the proposal explains who will benefit from the solution to the problem and in what ways.
- Literature review: the proposal is clearly related to a body of work and discusses the important publications in sufficient detail. The proposal gives proper citation for all significant concepts and ideas.
- Sound approach: the proposed solution is well-founded and has a reasonable chance of success.
- Originality: it is clear which concepts and ideas originate from the student and which are borrowed from the literature and from the discussions with the supervisor.
- Understanding of material: the student has a grasp of the scientific/mathematical concepts and of relevant algorithms at a sufficient level for carrying out the project.
- Capable of execution: the student has the necessary skills (programming or otherwise) to successfully complete the project in the allocated timeframe.
- Well-defined tasks and deliverables: each of the tasks in the plan is well defined with clear inputs and deliverables with a clear role to plan in delivering the final dissertation. The decomposition of the proposed problem/issue into subtasks is sound in the sense that it constitutes a plan that has a high chance of succeeding and producing deliverables that can be combined to support a passable MSc dissertation. It is also clear that there are no missing tasks - ensure adequate planning for writing up. Clear milestones are identified for each task.
- Project is feasible: the proposed work fits into the timescale for the dissertation. It is neither too big nor too small for the available time.
- Resources are attainable: the sources for any required equipment, funding and human effort are identified. It should be clear that the student will be capable of providing the necessary human effort.
- Evaluation: The proposal contains a concrete plan for evaluating the proposed work, including description of appropriate benchmark datasets, evaluation measures, gold-standard judgements and baselines to compare against.
Miscellaneous Resources
Here are some documents which may help you to formulate, write and present your proposal. They have been collected from various sources. Many of them are from the "Informatics Research Proposal" and "Informatics Research Review" courses taken by our regular MSc students, so
do not be confused by references to submission dates or other requirements for these courses. The requirements for our course are clearly stated above.
More generally, browse the
IRR and IRP wiki, which contains many more links to useful sources of advice, examples of good and bad proposals, and so on.
Ignore all references to tutor groups, submissions or other procedural details! These do not apply to PPar students. Instead, you should be discussing your proposal directly with your supervisor as it evolves.
Murray Cole, Informatics Forum room 1.18, ext. 505154