Friday, October 9, 2020

Do's and Don'ts of Applying to PhD Programs

 'Tis the season for applying to PhD programs and I'm giving a lot of the same advice.  This isn't comprehensive or official in any way, but here are some "Do's and Don'ts."

Don't cold email a bunch of professors generically asking if they are taking students this year, how does your resume look, etc.  Especially don't start your email with literally "Dear $name," (true story).

Do think carefully about the people you'd really like to be your advisor and request your mentors to introduce you if they can.  You can still cold email a few select people, but make sure the emails are short, easy to read, personal, and easily actionable.

Don't say you are interested in working with everyone in program.  If this is true, you probably need to narrow your interests.

Do pick a few relevant professors for each school and describe in your application why you'd like to work with them specifically.  This effort to tailor your applications is worth it.

Don't ask your letter writers to write for you last minute.

Do give at least 2 weeks notice, preferably more.  Make sure to pick letter writers who complement each other.  Consider your strategy of who is testifying to which of your skills and provide each with information on your application (for example, see my requirements).

Don't think that editing your written application materials isn't worth your time.  You need to write papers as a PhD student, so people pay attention to poor writing and grammatical errors.

Do have a few different people read your materials and give you feedback.

Don't assume poor grades, test scores, or holes in your application will go unnoticed.

Do provide explanations for any extenuating circumstances.

Don't focus only on one skill or attribute in your application; people are looking for the "whole package."

Do give evidence for the following:

  • Passion or desire to be in that particular program.  Why is this program/department the best fit for you?  Why not one of the other obvious choices?
  • Communication skills (e.g., good writing)
  • Relevant technical skills (e.g., statistical analysis, programming, ...)
  • Work Ethic.  Work Ethic.  Work Ethic.
  • Ability to complete projects.
  • Creativity.  Can you come up with your own ideas?  Will you be able to develop your own independent research agenda?

Wednesday, December 19, 2018

helping machine learning researchers understand the discipline of marketing

When I mention marketing to people in the machine learning community (and even scientists more broadly), some individuals become immediately disinterested or even express a degree of antipathy.  To me, this indicates they really have no understanding of the discipline.  The common perception is that marketing has to do with advertising and persuading people to buy things they don't want.

In reality, advertising and persuasion are only a part of marketing and the bit about getting people to buy things they don't want or need is bad marketing strategy.  Marketing is really more about identify what value you as a company can bring to a market and how best to profit from that value; one can see this by looking at the 3 C's and 4 P's framework that is foundational to the discipline:
  • Customer: What are the customer preferences?  What do they need?  What can they afford?
  • Competition: How are the customer needs already being met (or not) by others?
  • Company: How can we leverage our expertise and assets to give us a strategic advantage?
  • Product: What are we selling?  How does it meet the customer needs?
  • Promotion: How will customers find out about our product?  How can we help them understand that our product meets their needs?
  • Price: How much will we charge?  How will this impact the number of sales?
  • Place: How do we distribute our product?  Will our intended customers be able to access it?

The research side of marketing is inherently interdisciplinary, drawing from economics, psychology, statistics, and (more recently) computer science.  Drawing on such a wide pool of expertise makes sense—the problems marketing folks are tackling are big and messy with real-world impact.  Here are just a few examples of exciting areas for machine learning research in marketing:
  • Recommendation systems: an established application area in machine learning; looking at them through the marketing lens reveals new challenges and opportunities.
  • Impact of policies and information disclosure: when new policies (e.g., healthcare, nutrition, sustainability) are implemented, how does that impact the relevant, often very heterogeneous, markets?
  • Measurement models: how can we build interpretable machine learning models that can be used to identify some theoretical construct in marketing problems (e.g., customer sentiment or preference)?
  • Predicting demand: prediction problems fit neatly into the machine learning world; demand prediction is relevant for many markets, such as electricity, transportation (e.g., bike sharing, rental cars, flights), and selling physical objects (e.g., food, electronics).
I'll end this promotion of the marketing discipline with one thought.  Some very interesting marketing problems would benefit from expertise in machine learning.  Combine that with the need for more of the machine learning research community to ground the motivation for their work in real-world problems and there's a recipe for mutual benefit.

Friday, November 30, 2018

my new academic home

Changing affiliation from one academic discipline to another can be challenging, if not a bit risky, but I've decided to take the plunge: in July 2019, I'm starting as faculty at Duke's Fuqua School of Business in the Marketing area.  My background is in computer science, but academic disciplines are not orthogonal, and I think that my research interests fit both disciplines nicely.  By making the switch to marketing, I'm going to pushing myself and my research in ways that, while difficult at times, will be very beneficial because of the exposure to new perspectives.  I still love computer science, and hope to still collaborate with students and faculty in computer science (and marketing, statistics, and other disciplines), but here are some reasons I'm very excited to start as faculty in marketing.

Motivation is essential in marketing publications.  I've reviewed computer science papers (specifically machine learning) that try to get away with a single sentence of motivation; such an approach would be unheard-of in marketing.   Researchers in marketing are focusing on big, messy systems, and carve out research problems within the context of these complicated markets.  I believe the potential for direct real-world impact is much higher as a result.

Substantive claims are highly valued.  Methodological contributions are king in computer science, and focusing on predictive accuracy is common for evaluating machine learning methods.  While I will certainly still be working on methods, I am looking forward to make substantive contributions as well.  For example, I hope to understand the impact of algorithms on the markets in which they are deployed.

Journal over conference publications.  In computer science, researchers are typically expected to publish lots of shorter papers in conference proceedings; this encourages faster innovation, but it comes with the downside of less thorough publications.  In marketing, however, journal publications are the norm and that matches my research style better—I prefer taking my projects more slowly and carefully.  I'm looking forward to publishing longer papers.

Smaller research teams.  Marketing professors typically only mentor a handful of students at a time, compared to large research labs in computer science.  I like that I'll be able to give my students and their projects plenty of attention.

A better funding model.  The funding mechanisms at business schools center on tuition paid by Master's students, which means that time spent teaching those students is valued.  This model (vs. a grant-based model) is better aligned with how I'd like to allocate my time: I'll be able to focus on research, teaching, and mentoring students, without devoting large chunks of time to applying for grants.  To be clear, grant-writing is valuable process for research and I still plan to apply for grants, but it won't be a source of stress.  I hope this means I'll also be able to work faster and take bigger risks than I would under a grant-centered funding model.

I'm certain I will face challenges, but I am very excited to start this new adventure!