Communication and Power Networks: Architecture (Part I)

Adam took the courageous dive into the world of blogging about us at Caltech living in “the gap.”   I dare not commit to regular contributions as he admirably has, but have agreed to write a series of posts contrasting R&D for power and communication networks.  This is the first in this series, and my first-ever blog post!

Smart grid is in vogue these days, for good reasons.   As always, excitement garners people, ideas, and resources; but if not well-managed, can create disillusion that pushes the pendulum back.   It’d be impossible to predict how the current resurgence of interest in power systems R&D will play out, but the confluence of powerful forces will likely (and hopefully) drive dramatic advances in the coming decades.   We plan to chat about these in this space over the coming months.

Continue reading

A unusual submission trifecta

As many of you know, the Sigmetrics submission deadline is quickly approaching (Mon Dec 2nd).  The conference uses a double-blind submission policy and so I can’t post about what we are submitting yet, but we are certainly working hard on a few papers. We’ll see what happens, but it’s clear we’ll get at least one submitted.

Of course, next Monday is an awkwardly timed deadline, being the Monday after Thanksgiving, but it’s actually working out surprisingly well.  As a result of the holiday this week is a really quiet week on campus (no seminars, few meetings, few classes, etc.).  So, I have a lot more time than normal to focus on research, and have gotten a lot done.  The papers are in remarkably good shape given how far away the deadline is.  And, for the international students thanksgiving isn’t such a big deal, so they get a deadline week without much distraction.

But, what prompted this post wasn’t the deadline timing, it was the fact that I realized that these submissions will end an unusual trifecta for me.  This fall I’ll have submitted at least one paper to each of NSDI, STOC, and Sigmetrics; so a pure-systems conference, a pure-theory conference, and something in between.  (Maybe in a later post I’ll talk about the difference in writing papers for these communities — the presentation of results differs considerably.)

I’ve certainly submitted to each type of conference before, but never top-tier venues across the communities within such a small window of time.  (Of course who knows which, if any, will get accepted…)  It made me start to wonder: has anyone had papers in top-tier conferences of all three types of venues in the same year?  So, e.g., a paper in NSDI/Sigcomm, STOC/FOCS, and Sigmetrics/Performance.

I looked briefly and couldn’t find any in the last few years, but I could’ve easily missed someone.  I’ll be really interested to know if anyone is aware of this being accomplished…

The impact of awards on productivity

This post is brought on by an article I saw a while back about how winning awards changes people, which summarizes a recent paper by George Borjas and Kirk Doran on Prizes and Productivity: How winning the Fields medal affects scientific output.  The basic question in these articles is quite interesting:

What happens to researchers after they win major awards?

A pessimist might suggest that, upon receiving a major award, a researcher might  start to “slack off.”  It’s natural to think that after working hard for years to reach the pinnacle of their field, upon receiving the recognition, a certain loss of motivation might occur that could lead to a post-award slump… If we think of tenure as a major award, then it is certainly conventional wisdom that a post-tenure slump is common.

Basically, Borjas and Doran empirically tested this idea using data about the Fields medal, which is basically the Nobel prize of mathematics.  It is given to the top 2-4 mathematicians younger than 40 once every four years.

Continue reading

A call for postdocs

One of the things that I really enjoy about Caltech is the fact that we always have lots of amazing postdocs floating around…  For example, right now we’re lucky enough to have Siddharth Barman, Umang Bhaskar, and Yakov Babichenko doing joint CMI/SISL postdocs.

The thing I love about having postdocs around is how much I learn from them while they’re here.  Since we get postdocs in a wide variety of areas, they always come in as experts in some area that I wish I knew more about, and then, by the end of their stay, they’ve taught me a lot!  We actually formalize that process here at Caltech by having new postdocs give 3+ hour mini-courses the first term they arrive.  These are attended by nearly all of the related faculty and grad students…so by the end of the first few months, the postdocs have had a chance to train everyone enough to make them good candidates for collaboration!

All of that was a long intro to an advertisement…

We’re looking for new postdocs this year!

Continue reading

The tenuous ties between Sigmetrics and CS theory

While I publish in many venues, ACM Sigmetrics is certainly the closest thing I have to a home conference.  It is one of the few venues that can look at a mathematical paper and evaluate it both on its practical relevance and its technical depth — weighing the two together.  As a result, every conference ends up with a mixture of pure systems papers, pure theory papers, and a large number of hybrids with some degree of both systems and theory work.  Of course the balance between these types varies over the years, but in the decade I’ve been involved with the community, all types have maintained visibility.

Given that description, one might expect significant overlap of Sigmetrics with both more systemsy conferences like Sigcomm and NSDI, and more theoretical conferences like STOC and FOCS.  But, while this is true with NSDI and Sigcomm, there is virtually no overlap with STOC and FOCS.  One explanation for this could be that Sigmetrics isn’t really in the middle of theory and systems, it’s biased toward the systems side.  But, I don’t really think that’s true.  If anything, right now it’s biased towards the theoretical side of things.  To me, it seems to be more due to the fact that Sigmetrics uses a different sort of theory than the STOC and FOCS communities — there is much more applied probability and continuous math, because of a focus on stochastic performance bounds rather than worst-case guarantees.  (In the ’80s and early ’90s the theory side of Sigmetrics was dominated by queueing theory, which may be the historic reason for this difference.)

A shrinking divide?

Interestingly, it seems that the gap between the traditional theory CS theory community of STOC/FOCS and Sigmetrics has been shrinking in recent years due to the emergence of areas like “Algorithmic Game Theory” and “Network Economics.”  Questions at the intersection of CS, Networks, and Economics are drawing people from both the theory and the systems communities…  and this time, the techniques used by people across the communities overlap quite a bit.

Continue reading

A tale of heavy tails

For the past year, Bert Zwart, JK Nair and I have been working on a new book on heavy-tails.  The working title is “The fundamentals of heavy-tails: Properties, Emergence, and Identification.”

I’ll probably make a handful of posts about the book over the coming months, both on the topics we’re including and the process of writing, but I figure I’ll start by just giving an overview about why we’re even bothering to write such a book?

Why write a book on heavy-tails?

Continue reading

In praise of “neighborhood” workshops

Last week, we trekked from Caltech over to UCLA for the Southern California Network Economic and Game Theory (NEGT) workshop.    This was the fifth year for this event, which focuses on the intersection of CS, EE, and Economics. It rotates between USC, UCLA, and Caltech, and serves as a place where all the folks in the southern California area interested in topics at this intersection can get together and catch up.  This year, Mihaela van der Schaar and Bill Zame did a great job of hosting at UCLA.

I must say that I really love this sort of regional workshop — it’s one of my favorite events of the year.  We organize it with a bunch of faculty talks, pretty equally spilt across EE, CS, and Econ, and also pretty equally split across USC, UCLA, and Caltech.  Then, we bring in a few keynotes from outside of southern CA to round things out.  The quality of the talks is usually outstanding — and as a result, we tend to have 100+ people attend during the 2 days.  But, the best part is that the conversations that get started at the workshop really tend to extend throughout the year.  Since everyone is so close geographically, when common interests are discovered at the workshop, it tends to lead to multiple visits during the coming year… and since it’s been going on for five years now, we really have a community where folks in this area pretty much know everyone — regardless of the school and department they are from.  The workshop has really created a large and vibrant algorithmic game theory community here in the LA area, and one that is pretty unique in the sense that it truly bridges the three fields.  One thing that struck me this year was that it was often difficult to tell what department the speaker was from — by this point the topics & tools have really merged in our SoCal community.  I think this is really a consequence of the success of NEGT.

Continue reading