This section will focus on machine learning; specifically we will explore generative algorithms.
Week | Date | Topic | Paper Milestone |
---|---|---|---|
1 | Aug 29 | Introduction (PDF) | Introduction Survey |
Pair Presentations | |||
2 | Sept 5 | Alex, Munir, and Will → M. Jovanović | M0 - LaTeX Exercise |
3 | Sept 12 | Richard and Verrels → E. Bender | |
4 | Sept 19 | Arsum and David → P. Dhariwal | M1 - Topic Proposal |
5 | Sept 26 | Keyron and Stephen → A. Dosovitskiy | |
6 | Oct 3 | Hubery and Vivian → A. Vaswani | M2 - Bibliography |
7 | Oct 10 | EK and Grace → H. Jiang | |
Individual Presentations | |||
8 | Oct 17 | Alex → G. Bachmann; Munir → A. Golda | M3 - Annotated Bibliography |
9 | Oct 24 | Will → Y. Pandya; Hubery → J. Jumper | |
10 | Oct 31 | David → S. Chen; Richard → A. Kirillov | M4 - Intro + Outline |
11 | Nov 7 | Grace → H. Tatekawa; Rediet → N. Gillani | |
12 | Nov 14 | Arsum → D. Amodei; EK → J. Boucher | M5 - Full Paper Draft |
13 | Nov 21 | Stephen → J. Yoon; Keyron → D. Valevski; Verrels → C. Morrison | |
14 | Nov 28 | Thanksgiving | M6 - Final Paper |
15 | Dec 5 | Reading Day | |
16 | Dec 11 | Final Paper Presentations (Edmunds 114 9 to noon) |
List of papers for the first six weeks of paired presentations:
Here are some suggestions for your selected papers:
Papers chosen for the individual presentations:
See the home page for grading information.
You will find expectations for each category at these links:
When you are not presenting, you are expected to: read the listed article(s) and complete the questionnaire (due the day before the presentation).
All students are expected to participate in discussions after (and sometimes during) each presentation. I will update your participate grade on gradescope after each presentation.
For each presentation you must:
You’ll find additional presentation information in the following sections.
Here are some thing to consider before you present. We will discuss these items when we meet prior to your presentation.
Who is your audience? (peers, business, academics, investors, teaching, etc.). This will help you choose appropriate attire, mannerisms, formalisms, and engagements.
What methods will you use to calm your nerves? Bodies are funny; you can trick them into thinking that anxiety is excitement. We often produce excess adrenaline and epinephrine prior to presenting. These chemicals make us want to move. I recommend having a brisk walk prior to your presentation, learning some calming breathing techniques, and having a quick restroom break before you start.
How will you prepare your space? You should test your setup (laptop, dongle, projector) before you present. At a conference you should do so prior to the start of your session (not just before your present). I also like having something to drink on hand (water normally).
How will you reset your brain if you stumble? Have some idea of what you’ll do if you flub up a sentence. Maybe just take a longer drink of water to reset yourself.
What questions do you expect to receive? It will help you feel better prepared if you try to predict the questions you’ll receive.
How are you going to practice? The best forms of practice include some pressure and reviewing a recording of yourself.
How long will you each present? Try to stay on pace. Nobody likes it when a presenter rambles, and you’ll often get cutoff if you do so!
Here are the concepts on which I will provide feedback.
If you are looking for additional information, I recommend taking a look at these more formal/complete rubrics.
We will record all presentations, and you will review your recorded presentation with a CSWIM speaking partner after your presentation. You will also respond to the reflection prompts on gradescope. The key is to make mistakes, but not the same mistake twice! Receiving feedback is an emotional process. Don’t let your emotions get in the way of improving.
I do not have a preference for your presentation format. Common methods include:
The Think Fast, Talk Smart: The Podcast (here is a specific recommendation: Mindset Matters: How to Embrace the Benefits of Stress) is a good listen. You might as well start week 1!
If you are using slides, try to have a good mixture variety and mixture of text and graphics-only slides.
You should foreshadow and sign-post your presentations. But not necessarily by using an “outline” slide.
This is not a professional research talk. I would like to be informed, but it is also OK for you to also entertain.
It does not need to follow the same format/flow as the article(s) you present. A presentation is just an advertisement for the article.
Don’t memorize your “script,” but you can memorize transitions. If you over rehearse it will be difficult to recover from a mistake and it will come across as robotic.
Know how are you being “evaluated.” Is this a job talk (evaluated on ability to contribute to team)? Graduation keynote (people want to be inspired and enjoy the process)? For a grade (what is the rubric)? Informative (make sure you are credible)?
Presentations can come in any format. Here are a couple of great descriptions of how to craft a compelling presentation:
I HIGHLY recommend watching the 7 minute How to Make a Pecha Kucha video above. More information here Pecha Kucha 20x20: Discover Pecha Kucha presentations, stories, ideas, examples, and videos that will inspire.
Most presentations will follow this structure/outline:
Here are some tips for practicing your presentations:
Some quick tips from me:
Here are some tips from Some “Tips from a Pulitzer prizewinner” (US novelist Cormac McCarthy) written up by Van Savage and Pamela Yeh:
I recommend the following approach for finding research papers:
For other types of work (blog posts, etc.) use your normal searching skills (DDG, YouTube, etc.).
Keep the following in mind when reading any research article:
Who are the authors? The authors are the only people with the full context needed to understand a paper. Many authors are not native English writers, so don’t let lower-quality writing prevent you from understanding what could be an important takeaway.
What is the venue? Every venue will have its own requirements and expectations. Keep this in mind when thinking about what the authors left out (due to space requirements) or didn’t include (due to the media).
Who is the audience? You are likely not the intended audience for most research papers. The intended audience is probably some niche sub-field of researchers from around the world. Every group of researchers has its own jargon and informal (not codified) rules and conventions.
In general, I only gripe about grammar and style when reviewing (providing feedback prior to publication) an article, and not when reading an article or reviewing a research proposal. In the latter cases, do your best to look past writing characteristics (good or bad) and focus on contributions.
Once you find a research article (or research heavy blog post), I recommend the following technique:
In general, you only need to read in more detail if you need to understand something in more detail (for example, if you are trying to replicate their work). You should not expect to understand every part of a paper. You do not have the same background or context as the author(s). You will often want to read a paper several times if you are trying to recreate something the authors discovered.
You do not have to write your introduction in any specific format, but you might find Dr. Stirewalt’s 5-paragraph rule for writing Introductions helpful:
Introductory paragraph: What is the problem and why is it relevant to the audience attending THIS AUDIENCE? Moreover, why is the problem hard, and what is your solution? You must be brief here. This forces you to boil down your contribution to its bare essence and communicate it directly.
Background paragraph: Elaborate on why the problem is hard, critically examining prior work, trying to tease out one or two central shortcomings that your solution overcomes.
Transition paragraph: What keen insight did you apply to overcome the shortcomings of other approaches? Structure this paragraph like a syllogism: Whereas P and P = > Q, infer Q.
Details paragraph: What technical challenges did you have to overcome and what kinds of validation did you perform?
Assessment paragraph: Assess your results and briefly state the broadly interesting conclusions that these results support. This may only take a couple of sentences. I usually then follow these sentences by an optional overview of the structure of the paper with interleaved section callouts.
Most of you will decide to write a survey paper. So, what is a “survey paper”?
“A survey article … should provide a comprehensive review of developments in a selected area, covering its development from its inception to its current state and beyond, and illustrating its development through liberal citations from the literature.”
“A survey article assumes a general knowledge of the area; it emphasizes the classification of the existing literature, developing a perspective on the area, and evaluating trends.
Long Survey Paper: A paper that summarizes and organizes recent research results in a novel way that integrates and adds understanding to work in the field. A survey article assumes a general knowledge of the area; it emphasizes the classification of the existing literature, developing a perspective on the area, and evaluating trends.
Short Survey paper: Short survey papers are perspective pieces/essays that are well-focused and written by a leader in the field that usually presents a personal point of view critiquing widespread notions pertaining to a field. A perspective piece can be a review of a concept or a few related concepts.
Tutorial Paper: A paper that organises and introduces work in the field. A tutorial paper assumes its audience is inexpert; it emphasizes the basic concepts of the field and provides concrete example that embody these concepts.”
Here are two recent survey papers that you can use an example:
I will look for the following in your survey paper:
Survey papers will often include definitions and figures copied from other papers (cited, of course). As you read the articles you’ve accumulated, you should start organizing them into areas. The Research Paper Rubric will give you a good idea regarding how I will evaluate your report.
If you want more information, then it won’t hurt to skim through Procedures for Performing Systematic Reviews by Barbara Kitchenham.
Writing your technical report will be much easier if you start with an outline. Outlining will help you
An outline will typically follow this format:
Here is a more concrete example from Indeed’s advice on writing a research paper outline:
Here are three more sources for advice on how to outline a technical report:
We’ll perform peer reviews during two lecture periods. They will follow this format:
Note the following while reading your classmates’ reports (you should take notes on gradescope).
You can be brief in your note-taking.
Here is a senior thesis structure:
For this exercise you will use 𝐿𝐴𝑇𝐸𝑋 to create a “short autobiography.” Your autobiography can focus on whatever you’d like (it can be fiction if you’d like). You do not have any word or page count requirements, but you do need to include each of the features listed below.
Some general advice before you get started:
Your document must at minimum include the following:
\title
,
\author
, and \date
(see Your first
LaTeX document)
\abstract
(see formatting
abstracts)\section
s (see paragraphs
and
sections)~\ref
) each of your figures,
tables, and lists (see Cross
referencing sections, equations and floats).bib
file~\cite
) each reference in textThis GitHub Repository has quite a few tricks that you should consider if you are writing a larger document.
Here are some of my favorite (personally required) packages:
I use many of these in my proposal and cover letter templates.
Submit to gradescope.