Statistical Science

HOME | WHAT WE DO | WHO WE ARE | WHAT'S GOING ON | HOW TO JOIN

Many basic concepts developed originally for classical statistical physics, such as entropy and the renormalization group, wind up having much broader applications.  While our main focus is on electronic and atomic systems, group members have occasionally worked on problems from farther afield.  Some examples from the group's past work include the geometric theories of polymers and percolation, and analyses of localization effects on phonon transport; more recent examples include how amorphousness and topology affect the properties of photons (Figure 1) and how the macroscopic properties of knitted materials are affected by the underlying knit/purl topology (Figure 2).

Graph of experimental amorphous sample      Image of fabric fracture

Figure 1 (left): Graph describing an experimental photonic sample.  Figure 2 (right): Fabric fracture depends on knot topology.
Figures provided by Elizabeth Dresselhaus *23.

One can ask, in the many areas where physics touches on other fields, such as applied mathematics, computer science, biology, chemistry, statistics, and most of engineering, what physicists bring to the table: how is what physicists try to do different from the approach taken by other academics?  A personal opinion is that good physics seeks to be intellectually rigorous and falsifiable, even if it often falls short of mathematical rigor, and tries to isolate the key universal aspects of a problem rather than accumulating specific examples for their own sake ("stamp collecting", in Rutherford's famous phrase).  Of course, if a specific example is sufficiently important, then understanding its details may be a worthwhile exercise even for theoretical physicists.