Talks

In this talk, I will survey recent developments about the connection between neural networks and models of quantum mechanics and quantum field theory. Previous work has shown that the neural network - Gaussian process correspondence can be interpreted as the statement that large-width neural networks share some properties with free, or weakly interacting, quantum field theories (QFTs). Here I will focus on 1d QFTs, or models of quantum mechanics, where one has greater theoretical control. For instance, under mild assumptions, one can prove that any model of a quantum particle admits a representation as a neural network. Cherished features of quantum mechanics, such as uncertainty relations, emerge from specific architectural choices that are made to satisfy the axioms of quantum theory. Based on 2504.05462 with Jim Halverson.

The pseudo-Dirac Higgsino is one of the last surviving electroweak WIMPs. The LHC will not reach the 1.1 TeV target mass even with full luminosity and prospects for its indirect detection depend on a favorable dark matter density profile at the galactic center. Since it has only off-diagonal couplings at tree-level, its direct detection is possible only when the mass splitting is smaller than the initial center of mass kinetic energy. This direct detection loophole is actually more generic; going by the name 'inelastic dark matter'. In this talk I will talk about my recent efforts to reach mass splittings larger than what has been thought to be possible thus far, by invoking interesting astrophysics, elements from the end of the periodic table, and large volume neutrino detectors. 

Abstract

How do chemical reactions change when they’re run at temperatures a billion times colder than a Canadian winter? What can we learn when we have perfect quantum control of the reactants? Before answering these questions, we’ll discuss the fascinating techniques of laser cooling that allow us to cool atoms and molecules to within a few billionths of a degree above absolute zero. We’ll then look at how molecules prepared at such temperatures allow us to control chemical reactions at the quantum level, beginning to open a new understanding of chemistry and new possibilities for technologies of the future.

About the Speaker

Dr. Alan Jamison is an Assistant Professor at the University of Waterloo, jointly appointed to the Department of Physics and Astronomy and the Institute for Quantum Computing (IQC). He leads the Jamison Lab, which investigates ultracold atoms and molecules to explore quantum many-body physics, quantum chemistry, and quantum information science. Dr. Jamison earned his B.S. in Mathematics from the University of Central Florida in 2007, followed by an M.S. and Ph.D. in Physics from the University of Washington in 2008 and 2014, respectively.

After completing his Ph.D., he joined the group of Nobel Laureate Wolfgang Ketterle at the Massachusetts Institute of Technology (MIT) as a postdoctoral researcher. At the University of Waterloo, Dr. Jamison's research centers on using ultracold atoms and molecules to investigate complex quantum systems. His lab aims to achieve precise control over chemical reactions at ultracold temperatures, providing insights into quantum chemistry and enabling advancements in quantum computing and simulation.

Helen Hargreaves, MSW, RSW will present a workshop for PI Residents, providing information on the basics of Emotion Theory, how to assess ones own needs and communicate them. This presentation will particularly focus on Autistic and other neurodivergent ways of experiencing emotions and stress and how to better support neurodivergent team members in the workplace. Helen Hargreaves is a Neurodivergent Therapist with over 15 years experience workings with Neurodivergent clients. She is the Director of Rainbow Brain, a social work group practice that focuses on providing queer, trans and neurodivergent affirming therapy. Please note that this will be a 1.5 hour session with presentation and experiential components.

 In this talk, I will describe recent work on holographic correspondences in spacetimes which interpolate from anti-de Sitter space in the deep bulk to asymptotic regions which share some properties with flat space. Examples include the linear dilaton throat in the F1-NS5 solution and the NCOS decoupling limit of the D1-D5 system. In both examples, null geodesics take infinite coordinate time to reach the boundary, the causal structure resembles that of Minkowski space, and we can sensibly study radiation near future null infinity. These spacetimes are good solutions of string theory and thus might be considered candidates for a top-down sort of celestial holography.

The recent detection of a low frequency gravitational wave background by pulsar timing arrays provides solid evidence that the observable universe contains a population of in-spiralling supermassive black hole binaries. Such binaries likely form as a result of collisions between galaxies and can offer clues as to how black holes and galaxies grow over time. Moreover, since galactic centers are often densely populated by gas and stars, these systems are much more likely to be actively accreting and radiating compared to stellar mass-sized black hole binaries. This makes them promising candidates for multi-messenger detection (combining electromagnetic and gravitational wave information), particularly when LISA comes online or as pulsar timing arrays improve their sensitivity. In order to facilitate this goal, it is important that, in addition to predictions for the possible gravitational wave signals from numerical relativity, we also develop predictions for electromagnetic signals from simulations of black hole binary accretion. In this talk I will present some of our recent results from 3D general relativistic magnetohydrodynamic simulations that utilize a strong-field approximation to the in-spiralling spacetime metric. Specifically, I will showcase several possible distinctive electromagnetic signals from black hole binaries, including quasi-periodic emission, jet precession, and two processes analogous to flaring activity frequently observed in the outer layer of the Sun.

We show that specific exponential integrals serve as generating functions of labeled edge-colored graphs. Based on this, we derive asymptotics for the number of edge-colored graphs with arbitrary weights assigned to different vertex structures. The asymptotic behavior is governed by the critical points of a polynomial. As an application, we discuss the Ising model on a random graph and show how its phase transitions arise from our formula.

We explore the interplay of quantum computing and machine learning to advance experimental protocols for observing measurement-induced phase transitions (MIPT) in quantum devices. In particular, we focus on trapped ion monitored circuits and apply the cross entropy benchmark recently introduced by [Li et al., Phys. Rev. Lett. 130, 220404 (2023)], which can mitigate the postselection problem. By doing so, we reduce the number of projective measurements -- the sample complexity required per random circuit realization, which is a critical limiting resource in real devices. Since these projective measurement outcomes form a classical probability distribution, they are suitable for learning with a standard machine learning generative model. In this work, we use a recurrent neural network (RNN) to learn a representation of the measurement record for a native trapped-ion MIPT, and show that using this generative model can substantially reduce the number of measurements required to accurately estimate the cross entropy. This illustrates the potential of combining quantum computing and machine learning to overcome practical challenges in realizing quantum experiments.