Scribd
Upload
108 views30 pages

How To Be A Good PHD Student: Elena Dubrova Ecs/Ict/Kth

This document outlines the objectives and content of a course on how to be a good PhD student. The course aims to improve students' ability to present research results through writing papers, giving talks, and writing proposals. It will also cover research ethics. The course content is divided into introductory, writing papers, giving talks, writing proposals, and research ethics sections. Students will apply the techniques learned by writing a paper, presenting it, and applying for a travel grant. The course will run over an academic year with lectures every other week.

Uploaded by

Nasri Atkinson
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
108 views30 pages

How To Be A Good PHD Student: Elena Dubrova Ecs/Ict/Kth

This document outlines the objectives and content of a course on how to be a good PhD student. The course aims to improve students' ability to present research results through writing papers, giving talks, and writing proposals. It will also cover research ethics. The course content is divided into introductory, writing papers, giving talks, writing proposals, and research ethics sections. Students will apply the techniques learned by writing a paper, presenting it, and applying for a travel grant. The course will run over an academic year with lectures every other week.

Uploaded by

Nasri Atkinson
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 30

How to be a good PhD student

Elena Dubrova
ECS/ICT/KTH

Objectives of the course


To improve your the ability to present your
research results, i.e. to learn how to:
1. Write a good research paper
2. Give a good technical presentation
3. Write a successful project proposal

To inform you about research ethics

p. 2 - FIL3001 The Art of Doctoral Research

Content of the course


The course consists of an introductory part, 3
main parts focusing on the objectives described
on the previous slide, and the research ethics
part
Introductory part will cover major phases of a
PhD study:
Background study, Literature survey, Detailed problem
formulation, Execution, Publishing

p. 3 - FIL3001 The Art of Doctoral Research

First part: Writing papers


Gives general guidelines and specific
suggestions about writing technical papers
We will study major sections of a paper:
Introduction, Related work, Research presentation,
Experimental results, Conclusion, Future work

We will cover different paper categories:


Concept paper, paper presenting experimental
results, theoretical work, methodology, etc.

p. 4 - FIL3001 The Art of Doctoral Research

Second part: Giving talks


Covers techniques for giving technical talks
We will review common ways to format the slides
and to convey the information to the audience
We will discuss typical ways to organize of a
conference talk:
Problem description, Motivation, Background,
Previous Work, Novelty and Contribution,
Experimental results, Conclusion and Future Work

p. 5 - FIL3001 The Art of Doctoral Research

Third part: Writing proposals


Provides information about sources of external
funding and guidelines in writing project
applications
We will learn how to organize a comprehensive
funding application and to present it to the
outside world in a format that is recognized and
accepted

p. 6 - FIL3001 The Art of Doctoral Research

Last Part: Research Ethics


This part covers the main aspects of research
ethics, including the design and implementation of
research involving human and animal
experimentation, various aspects of scientific
misconduct (such as fraud, fabrication of data and
plagiarism), as well as codes of ethics and
professional conduct

p. 7 - FIL3001 The Art of Doctoral Research

Theory vs practice
The theoretical part of the course will be
complemented by practical exercises
Each student will apply the techniques learned in
the course to write a conference paper in his/her
area of research, to present this paper in the
class, and to apply for a travel grant (e.g.
supporting a conference trip or a research visit)

p. 8 - FIL3001 The Art of Doctoral Research

Course structure
Length:
one academic year Sept 2016 June 2017

Lectures
Typically every other week, on Thursday 10-12, see
course web page for the schedule:
https://people.kth.se/~dubrova/coursePhD.html

p. 9 - FIL3001 The Art of Doctoral Research

Lecture 1: Introductory Part

Introductory part
Major phases of a PhD study are:

Background study
Literature survey
Detailed problem formulation
Execution
Publishing

p. 11 - FIL3001 The Art of Doctoral Research

Background study
The goal is to understand needs in your area in
order to focus your PhD research on an
important problem
This task is hardest of all

p. 12 - FIL3001 The Art of Doctoral Research

Important to know
An essential feature of a good paper, talk, or
proposal is:
The ratio information/noise is maximized

Why so many papers, talk, and proposals are


bad?
Because the author does not have a clear picture of
the goal, objectives, and results of the presented
research, as well as its place in the world

p. 13 - FIL3001 The Art of Doctoral Research

Basic Terminology: Goal


Goal should explain what the core problem is
and why it is important
There should be only one goal
The goal should be connected to the vision for
development
It should be difficult or impossible to measure the
accomplishment of the goal using measurable
indicators, but it should be possible to prove its merit
and contribution to the vision

p. 14 - FIL3001 The Art of Doctoral Research

Basic terminology: Objectives


Objectives should address the core problem in
terms of the benefits to be received by the target
group
Objectives provide a more detailed breakthrough
of the project goal

p. 15 - FIL3001 The Art of Doctoral Research

Basic terminology: Results


Results describe the services or products to be
delivered to the intended target group
To ensure relevant results, one should have
correctly identified the group's needs
It should be possible to measure results through
the use of objective indicators

p. 16 - FIL3001 The Art of Doctoral Research

Basic Terminology: Indicators


Indicators are used to evaluate the success

Quality
Target group (who?)
Place (where?)
Quantity (how much?)
Time (by when?)

p. 17 - FIL3001 The Art of Doctoral Research

Example I
White paper on Physical Synthesis for Power
Under Process Variation of J. Cong
Structure:

Motivation
Objectives
Technical approaches
Anticipated results

p. 18 - FIL3001 The Art of Doctoral Research

Motivation
The most significant progress in EDA in the past ten years is arguably the
development of physical synthesis technology and its wide adoption by the IC
design industry today. However, the existing physical synthesis technology has
several limitations:
(1) It was originally developed to address the timing closure problem, thus, did
not give sufficient consideration of power optimization. Although a number of
power optimization techniques have been added in many physical synthesis
flows (such as cell sizing, multiple Vt and Vdd selections), there is lack of
general and efficient algorithmic framework to consider and balance all
available power optimization opportunities, especially in connection with
placement.
(2) It did not consider increasing process variations in nanometer designs,
therefore, the results by existing physical optimization algorithms may not work
under all timing corners. A considerable safe margin is added to guard-band
the statistical variation, resulting in a sizable waste of power, area, and
performance.
p. 19 - FIL3001 The Art of Doctoral Research

Objectives
(a) Develop a unified and efficient mathematical foundation and algorithmic
framework for physical synthesis to support power optimization guided by both
physical locality and an evolving netlist structure. For example, the placement
engine should support multi-Vdd islands, clock gating and power gating, where
physical locality has a big impact on optimization quality. But it should also
support cell sizing, buffering, logic structuring, where an evolving netlist and
thus changing logic density need to be considered, during placement, for
correct power optimization.
(b) Develop efficient theory and algorithms to support statistical optimization under
process variation so that the physical synthesis results satisfy all timing
constraints under all timing corners or achieve the required timing-yield
constraints under given probability density functions (PDFs).
(c) Consider the efficient use of multi-core CPUs (as they become widely
available) to further improve the runtime efficiency for coping with the high
complexity of the proposed problem.
p. 20 - FIL3001 The Art of Doctoral Research

Anticipated results
Anticipated results of the proposed research include
technical reports, published papers in major EDA
conferences and journals, and a software prototype of a
novel physical synthesis flow for power optimization under
process variation. Our research group has a strong track
record in delivering the results and transferring the
technology to the SRC companies our latest SRC project
on placement (ending June06) promised a Moores law
generation reduction of wirelength (30%) and we already
exceeded that goal considerably [Chan05].

p. 21 - FIL3001 The Art of Doctoral Research

Example II

p. 22 - FIL3001 The Art of Doctoral Research

Good source of needs - ITRS


International Technology Roadmap for
Semiconductors forecasts needs of
semiconductor industry in the near term (3 years)
and long term (up to 10 years)
Design and test challenges, emerging devices,
assembly and packaging, etc.

Home page at
http://www.itrs.net/

p. 23 - FIL3001 The Art of Doctoral Research

Example of design challenges from ITRS


The following 5 problems are considered to be
the largest future challenges:
1.
2.
3.
4.
5.

Design productivity
Power consumption
Manufacturability
Interference
Error tolerance

p. 24 - FIL3001 The Art of Doctoral Research

Example of challenges, cont.


We are facing two types of complexity:
Silicon complexity refers to the impact of process
scaling and the introduction of new materials or
device/interconnect architectures
System complexity refers to exponentially increasing
transistor counts enabled by smaller feature sizes and
urged by consumer demand for increased
functionality, lower cost and shorter time-to-market

p. 25 - FIL3001 The Art of Doctoral Research

Silicon complexity implications


Many previously ignorable phenomena now have
great impact on design correctness and value:
Non-ideal scaling of device parasitics and
supply/threshold voltages
Coupled high-frequency devices and interconnects
Manufacturing variability
Decreased reliability

p. 26 - FIL3001 The Art of Doctoral Research

Silicon complexity implications, cont.


Silicon complexity places long-standing
paradigms at risk:
System-wide synchronization becomes infeasible due
to power limits and the cost of robustness under
manufacturing variability
The CMOS transistor becomes subject to ever-larger
statistical variabilities in its behavior
Fabrication of chips with 100% working transistors
and interconnects becomes prohibitively expensive

p. 27 - FIL3001 The Art of Doctoral Research

System complexity implications


Implied challenges include:
Design specification and verification become
extremely complex
Need for support for hierarchical design,
heterogeneous Soc integration, etc.
Trade-offs must be made between all aspects of value
or quality, and all aspects of cost

p. 28 - FIL3001 The Art of Doctoral Research

First assignment
The purpose is to persuade you that you will be
able to produce a good paper/talk/proposal with
a maxmum information/noise ratio only if:
1. You understand the core problem which you are
addressing, and
2. You are aware of the needs for solving this problem,
and
3. You have a clear picture of the goal, objectives, and
expected results of your research

p. 29 - FIL3001 The Art of Doctoral Research

First assignment, cont.


Find one problem which you consider important
For example, take the project on which you
are working for you PhD
Define what would be you goal, objectives,
expected results and indicators of success
Present your answer in the form of a diagram
similar to the example on p.21
The assignment is due on Sunday, Oct. 2nd.
Send it by email ([email protected]) as a pdf file
with subject assignment 1 FIL3001
p. 30 - FIL3001 The Art of Doctoral Research

You might also like