Talks

Machine learning has led to recent advancements in image processing, language translation, finance, robotics, musical and visual arts, and medical diagnosis. In this session, we will explore how machine learning can be applied within fields of physics. We will introduce fundamental concepts in machine learning such a neural networks and supervised vs. unsupervised learning, and then proceed to learn to use tools from Python's TensorFlow library.

Bring your laptop. You can attend remotely via Zoom <https://zoom.us/j/154009181>.

Ghostly neutrino particles continue to bring surprises to fundamental physics, from their existence to the phenomenon of neutrino oscillation which implies that their masses are nonzero. Their exact masses, among the most curious unknowns beyond the Standard Model of particle physics, can soon be probed by the joint analysis of upcoming cosmological surveys including LSST, Euclid, WFIRST, Simons Observatory, and CMB-S4. In this talk, I will first discuss ongoing work studying the effects of massive neutrinos. I will then turn the focus to my major efforts of modeling the challenging nonlinear regime of cosmic structures (<10 Mpc) where neutrino effects are the strongest. Finally, I will draw a roadmap to pin down the neutrino mass over the next decade.

We give a quantum speedup for solving the canonical semidefinite programming relaxation for binary quadratic optimization. The class of relaxations for combinatorial optimization has so far eluded quantum speedups. Our methods combine ideas from quantum Gibbs sampling and matrix exponent updates. A de-quantization of the algorithm also leads to a faster classical solver. For generic instances, our quantum solver gives a nearly quadratic speedup over state-of-the-art algorithms.
This is joint work with Fernando Brandao (Caltech) and Daniel Stilck Franca (QMATH, Copenhagen).

The asymptotic symmetry charge algebra of certain class of spacetimes could have a nontrivial central extension, which measures the non-equivariance of the charges of the large gauge transformations. The Cardy formula of the dual CFT has been famously used to derive black hole entropy. However, it remains obscure from the point of view of gravity why such a constant on the gravitational phase space could encode the information about the density of black hole micro-states, and what the degrees of freedom accounting for the black hole entropy truly are. I will discuss the ongoing efforts of understanding these questions in the covariant phase space formalism.

There is a close relationship between derived loop spaces, a geometric object, and Hochschild homology, a categorical invariant, made possible by derived algebraic geometry, thus allowing for both intuitive insights and new computational tools.  In the case of a quotient stack, we discuss a "Jordan decomposition" of loops which is made precise by an equivariant localization result.  We also discuss an Atiyah-Segal completion theorem which relates completed periodic cyclic homology to Betti cohomology.