Upcoming Talks

Lean is a functional programming language that is designed to be used for theorem proving. Functions in lean which compile can be directly mapped to theorems, making the lean compiler a powerful tool for theorem verification. I will describe the basics of of lean, and build up towards the definitions and structures needed in quantum information. The particular theorem we will work towards is the security guarantee of the bb84 quantum key distribution protocol. The seminar is intended to be accessible for physicists with no prior knowledge of lean.

BSc: Victoria university of Wellington, New Zealand. MSc+PhD: Uppsala University, Sweden Postdoc: Humboldt University, Berlin, Germany Postdoc: MIT

Controlling nonlinear multimode states is a long-standing challenge in physics. While systems with high nonlinearities, such as superconducting circuits and acousto-optic systems, have successfully demonstrated precise control over quadrature-dependent behavior (such as quantum squeezing, quadrature non-reciprocity, and bosonic Kitaev chains), integrated photonics faces a unique hurdle. Specifically, the inherently weak nonlinear interactions in dielectric materials have limited the full potential of optical platforms. However, recent advances in fabricating nonlinear micro-resonators with nanometric features now allow for the enhancement of all-optical interactions, necessitating new approaches to generating, controlling, and measuring light. In this seminar, I will discuss our recent results in observing, controlling, and programming multimode quadrature-dependent Hamiltonians to enable new on-chip functionalities. I will begin by showcasing our advancements in developing integrated microresonators in thin-film 4H-Silicon Carbide. This innovation enables nonlinear photonics, quantum optics, and collective quantum emitter excitations on a single platform. Following this, I will present the experimental demonstration of a fully tunable optical dimer that exhibits complete quadrature-dependent non-reciprocity/isolation, opening the door to enhanced light-matter interactions and sensing. Finally, I will discuss the observation and control of quadrature-dependent dynamics that naturally emerge in Kerr micro-combs. Our work paves the way toward new regimes of light-matter interactions on scalable photonic microchips, with transformative implications for both fundamental physics and quantum applications.

Dr. Eran Lustig is an Assistant Professor in the Faculty of Electrical and Computer Engineering at the Technion. His primary research interests lie in optical physics and engineering, with a focus on nonlinear, multimode, topological, and time-dependent optics on various optical platforms. Dr. Lustig earned his PhD in Physics from the Technion and recently completed his postdoctoral studies at the Ginzton Laboratory at Stanford University. He is currently a Seiden Fellow in Nanotechnology and Optoelectronics and was the recipient of the IPS Asher Peres Award for Outstanding Experimental Student.

Modern edge devices, such as cameras, drones, and Internet-of-Things nodes, rely on deep learning to enable a wide range of intelligent applications. However, this deep learning inference is usually disaggregated: the model is stored on the cloud, while the inputs/outputs are obtained/required on the edge. To this end, we present a novel disaggregated computing architecture for wireless edge networks with two key innovations: disaggregated model access via over-the-air wireless broadcasting for simultaneous inference on multiple edge devices, and in-physics computation of general matrix-vector multiplications directly at radio frequency driven by a single frequency mixer. Using an experimental software-defined radio platform, it achieves 95.7\% image classification accuracy with ultra-low energy consumption of more than two orders of magnitude improvement compared to traditional digital computing.

Zhihui Gao is a final-year Ph.D. Student co-advised by Prof. Tingjun Chen and Prof. Yiran Chen at Duke University. His research interest is the next generation of network systems, machine learning acceleration, and cyber-physical systems. His work has been published in top venues such as Science Advances, ACM MobiCom, ACM MobiHoc, and ACM/IEEE IPSN.