With the advent of gravitational-wave astronomy, techniques to extend the reach of gravitational- wave detectors are desired. In addition to the stellar mass black hole and neutron star mergers already detected, many more are below the surface of the noise, available for detection if the noise is reduced enough. One method for noise reduction applies machine learning algorithms to gravitational-wave detector data and auxiliary channels on-site to reduce the noise in the time-series due to instrumental artifacts. Given realistic assumptions about coupling mechanisms, we are able to reduce the noise floor, leading to detector sensitivity improvements. This framework is generic enough to subtract both linear and non-linear coupling mechanisms, and learn about the mechanisms which are not currently understood to be limiting detector sensitivities. We discuss lessons learned and how this work can be generalized to other time series regression analyses in all areas of science.

]]>Ising model is a classical model for ferromagnetism in statistical mechanics. In a joint work with Pavel Galashin we completely describe by inequalities the set of boundary correlation matrices of planar Ising networks embedded in a disk. Specifically, we give a simple bijection between such correlation matrices and points in the totally nonnegative part of the orthogonal Grassmannian, which has been introduced recently in the study of the scattering amplitudes of ABJM theory. Under our correspondence, Kramers--Wannier's high/low temperature duality transforms into the cyclic symmetry of the Grassmannian. We also show that the edge parameters of the Ising model for reduced networks can be uniquely recovered from boundary correlations, solving the inverse problem.

]]>When left unobserved, many-body quantum systems tend to evolve toward states of higher entanglement. Making a measurement, on the other hand, tends to reduce the amount of entanglement in a many-body system by collapsing one of its degrees of freedom. In this talk I discuss what happens when a many-body quantum system undergoes unitary evolution that is punctuated by a finite rate of projective measurements. Using numerical simulations and theoretical scaling arguments, we show that for a 1D spin chain there is a critical measurement rate separating two dynamical phases. At low measurement rate, the entanglement grows linearly with time, producing a volume-law entangled state at long times. When the measurement rate is higher than the critical value, however, the entanglement saturates to a constant as a function of time, leading to area-law entanglement. We map the dynamical behavior of the entanglement onto a problem of classical percolation, which allows us to obtain the critical scaling behavior near the transition. I briefly discuss generalizations of our result to higher dimensions, and its implications for the difficulty of simulating quantum systems on classical computers.

]]>This is public portion of Ms. Straub's Thesis Defense. Her advisors are Leon Hsu and Ken Heller.

Over the past half century, researchers and curriculum developers studying physics education have created dozens of innovative curricula and educational tools, broadly referred to as research-based instructional strategies (RBIS), to fit almost any classroom situation. These include cooperative problem solving (Heller & Hollabaugh, 1992; Heller, Keith, & Anderson, 1992), Physics By Inquiry (McDermott, Shaffer, & Rosenquist, 1996), Investigative Science Learning Environment (ISLE) (Etkina & Van Heuvelen, 2007), Studio Physics (Cummings et al., 1999), and Peer Instruction (Crouch & Mazur, 2001) among others. However, the rate of adoption of RBIS remains relatively low. A national survey of post-secondary physics instructors in 2012 showed that only half of physics instructors have ever implemented any RBIS in their classrooms, and many of them ceased to do so after implementation difficulties (Henderson & Dancy). Why arenât these effective strategies being implemented at larger rates? Part of removing barriers to RBIS adoption may be understanding what instructors believe about how students learn.

In order to answer a small portion of this question, I studied physics instructorsâ beliefs about homework. This study is taken up in two parts. First, I analyzed 25 interviews with physics instructors from various types of institutions in Minnesota (Yerushalmi et al., 2007; Henderson, et al., 2007). Second, I used the themes from the interview analysis to create a survey, which was then sent to physics instructors in the state of Minnesota. Using both the interview analysis and the survey responses, I created an empirical model of physics instructorsâ beliefs about homework. There were four main results. First, there is agreement that the goals of doing homework are to learn problem solving and physics principles. Second, homework is seen as necessary for learning physics by a strong majority of instructors, but it is not seen as sufficient for learning. Third, there is a limited number of tasks or actions that instructors believe that students should do while they are solving problems. Fourth, there is evidence that physics instructors fall onto a continuum of beliefs regarding how students should approach solving problems on their homework. On one end of this continuum, instructors believe students should follow an algorithmic process that includes the steps to solving any problem. On the other end of the continuum, instructors believe students should have a more open approach to solving problems where they consider all the tools and principles available to them in order to make decisions about how to solve a problem. These results can inform creators of curriculum and professional development as they try to reach out and connect with instructors and perhaps change their beliefs and practice.

]]>New and improved observational facilities are sampling the night sky with unprecedented temporal cadence and sensitivity across the electromagnetic spectrum. This exercise led to the discovery of new types of astronomical transients and revolutionized our understanding of phenomena that we thought we already knew. In this talk I will review some very recent developments in the field that resulted from the capability to acquire a true panchromatic view of the most extreme stellar deaths in nature.

]]>