Seminars and Colloquia by Series

Braess's Paradox in Expanders

Series
Other Talks
Time
Monday, October 8, 2012 - 13:05 for 1 hour (actually 50 minutes)
Location
Klaus 1116W
Speaker
Stephen YoungUniversity of Louisville, Kentucky
Expander graphs are known to facilitate effective routing and most real-world networks have expansion properties. At the other extreme, it has been shown that in some special graphs, removing certain edges can lead to more efficient routing. This phenomenon is known as Braess¹s paradox and is usually regarded as a rare event. In contrast to what one might expect, we show that Braess¹s paradox is ubiquitous in expander graphs. Specifically, we prove that Braess¹s paradox occurs in a large class of expander graphs with continuous convex latency functions. Our results extend previous work which held only when the graph was both denser and random and for random linear latency functions. We identify deterministic sufficient conditions for a graph with as few as a linear number of edges, such that Braess¹s Paradox almost always occurs, with respect to a general family of random latency functions. Joint work with Fan Chung and Wenbo Zhao. (* Note that this is an ARC/Theory Seminar and is in Klaus 1116W *)

On the KLR conjecture in sparse random graphs

Series
Other Talks
Time
Friday, October 5, 2012 - 16:00 for 1 hour (actually 50 minutes)
Location
**Emory University**, Mathematics and Science Center, Rm W201
Speaker
Mathias SchachtMath, University of Hamburg, Germany
(**This is at Emory and is a joint Emory - Georgia Tech Combinatorics Seminar. **) The KLR conjecture of Kohayakawa, Luczak, and Rödl is a statement that allows one to prove that asymptotically almost surely all subgraphs of the random graph G(n,p) satisfy an embedding lemma which complements the sparse regularity lemma of Kohayakawa and Rödl. We prove a variant of this conjecture which is sufficient for most applications to random graphs. In particular, our result implies a number of recent probabilistic threshold results. We also discuss several further applications. This joint work with Conlon, Gowers, and Samotij.

Nanoengineered Surfaces: Transport Phenomena and Energy Applications

Series
Other Talks
Time
Friday, October 5, 2012 - 11:00 for 1 hour (actually 50 minutes)
Location
MRDC, Room 4211
Speaker
Evelyn WangDepartment of Mechanical Engineering, MIT

Please Note: Host: David Hu. Refreshments will be served. Speaker's Bio

Nanoengineered surfaces offer new possibilities to manipulate fluidic and thermal transport processes for a variety of applications including lab-on-a-chip, thermal management, and energy conversion systems. In particular, nanostructures on these surfaces can be harnessed to achieve superhydrophilicity and superhydrophobicity, as well as to control liquid spreading, droplet wetting, and bubble dynamics. In this talk, I will discuss fundamental studies of droplet and bubble behavior on nanoengineered surfaces, and the effect of such fluid-structure interactions on boiling and condensation heat transfer. Micro, nano, and hierarchical structured arrays were fabricated using various techniques to create superhydrophilic and superhydrophobic surfaces with unique transport properties. In pool boiling, a critical heat flux >200W/cm2 was achieved with a surface roughness of ~6. We developed a model that explains the role of surface roughness on critical heat flux enhancement, which shows good agreement with experiments. In dropwise condensation, we elucidated the importance of structure length scale and droplet nucleation density on achieving the desired droplet morphology for heat transfer enhancement. Accordingly, with functionalized copper oxide nanostructures, we demonstrated a 20% higher heat transfer coefficient compared to that of state-of-the-art dropwise condensing copper surfaces. These studies provide insights into the complex physical processes underlying fluid-nanostructure interactions. Furthermore, this work shows significant potential for the development and integration of nanoengineered surfaces to advance next generation thermal and energy systems.

Discrete Mathematical Biology Working Seminar

Series
Other Talks
Time
Tuesday, October 2, 2012 - 11:00 for 1 hour (actually 50 minutes)
Location
Skiles 114
Speaker
David Murrugarra Georgia Tech
A discussion of the paper "Probabilistic Boolean networks: a rule-based uncertainty model for gene regulatory networks" by Shmulevich et al.

Discrete Mathematical Biology Working Seminar

Series
Other Talks
Time
Tuesday, September 25, 2012 - 10:00 for 1 hour (actually 50 minutes)
Location
Skiles 114
Speaker
Will PerkinsGeorgia Tech
Further discussion of co-transcriptional RNA folding, and the potential for trap models to capture these dynamics.

Discrete Mathematical Biology Working Seminar

Series
Other Talks
Time
Tuesday, September 18, 2012 - 10:00 for 1 hour (actually 50 minutes)
Location
Skiles 114
Speaker
Will PerkinsGeorgia Tech
We will continue discussing co-transcriptional RNA folding, and the potential for trap models to capture these dynamics.

Discrete Mathematical Biology Working Seminar

Series
Other Talks
Time
Tuesday, September 11, 2012 - 10:00 for 1 hour (actually 50 minutes)
Location
Skiles 114
Speaker
Will PerkinsGeorgia Tech
We will discuss how best to model and predict the co-transcriptional effects of RNA folding. That is, using the fact that the RNA molecule begins folding as the sequence is still being transcribed, can we find better predictions for the secondary structure? And what is a good mathematical model for the process?

Algebraic Geometry for Applications (IMA PI summer program; June 18th - July 6th)

Series
Other Talks
Time
Monday, June 18, 2012 - 09:30 for 8 hours (full day)
Location
Klaus 1116
Speaker
Greg Blekherman, Anton Leykin, and Josephine YuGeorgia Tech
This is a summer school (June 18th - July 6th) in computational algebraic geometry intended for graduate students, however, everyone is welcome to attend. For details and schedule see aga.gatech.edu. The first day's schedule has been slightly altered; we will give introductory lectures at 9:30 (Anton Leykin -- Computer Algebra and Numerical Algebraic Geometry), 11:30 (Greg Blekherman -- Convexity), and 2:00 (Josephine Yu -- Tropical Geometry).

ACO/CS Theory Seminar - Solving maximum flows in O(nm) time, and less

Series
Other Talks
Time
Wednesday, May 23, 2012 - 13:00 for 1 hour (actually 50 minutes)
Location
Klaus 1116W
Speaker
Jim OrlinMIT Sloan Management
Over the past 30 years, researchers have developed successively faster algorithms for the maximum flow problem. The best strongly polynomial time algorithms have come very close to O(nm) time. Many researchers have conjectured that O(nm) time is the "true" worst case running time. We resolve the issue in two ways. First, we show how to solve the max flow problem in O(nm) time. Second, we show that the running time is even faster if m = O(n). In this case, the running time is O(n^2/log n).

Pages