arXiv subjects

Zachary Smith

Publications and source records attributed to Zachary Smith.

7 recordsLinked to original sources

arXiv2025

Telecommunications fiber-optic and free-space quantum local area networks at the Air Force Research Laboratory

As quantum computing, sensing, timing, and networking technologies mature, quantum network testbeds are being deployed across the United States and around the world. To support the Air Force Research Laboratory (AFRL)'s mission of building heterogeneous quantum networks, we report on the development of Quantum Local Area Networks (QLANs) operating at telecommunications-band frequencies. The multi-node, reconfigurable QLANs include deployed optical fiber and free-space links connected to pristine laboratory environments and rugged outdoor test facilities. Each QLAN is tailored to distinct operating conditions and use cases, with unique environmental characteristics and capabilities. We present network topologies and in-depth link characterization data for three such networks. Using photonic integrated circuit-based sources of entangled photons, we demonstrate entanglement distribution of time-energy Bell states across deployed fiber in a wooded environment. The high quality of the entanglement is confirmed by a Clauser-Horne-Shimony-Holt inequality violation of $S=2.717$, approaching the theoretical maximum of $S=2.828$. We conclude with a discussion of future work aimed at expanding QLAN functionality and enabling entanglement distribution between heterogeneous matter-based quantum systems, including superconducting qubits and trapped ions. These results underscore the practical viability of field-deployable, qubit-agnostic quantum network infrastructure.

quant-ph↗

arXiv2025

Noise-Aware Entanglement Generation Protocols for Superconducting Qubits with Impedance-Matched FBAR Transducers

Connecting superconducting quantum processors to telecommunications-wavelength quantum networks is critically necessary to enable distributed quantum computing, secure communications, and other applications. Optically-mediated entanglement heralding protocols offer a near-term solution that can succeed with imperfect components, including sub-unity efficiency microwave-optical quantum transducers. The viability and performance of these protocols relies heavily on the properties of the transducers used: the conversion efficiency, resonator lifetimes, and added noise in the transducer directly influence the achievable entanglement generation rate and fidelity of an entanglement generation protocol. Here, we use an extended Butterworth-van Dyke (BVD) model to optimize the conversion efficiency and added noise of a Thin Film Bulk Acoustic Resonator (FBAR) piezo-optomechanical transducer. We use the outputs from this model to calculate the fidelity of one-photon and two-photon entanglement heralding protocols in a variety of operating regimes. For transducers with matching circuits designed to either minimize the added noise or maximize conversion efficiency, we theoretically estimate that entanglement generation rates of greater than $160\;\mathrm{kHz}$ can be achieved at moderate pump powers with fidelities of $>90\%$. This is the first time a BVD equivalent circuit model is used to both optimize the performance of an FBAR transducer and to directly inform the design and implementation of an entanglement generation protocol. These results can be applied in the near term to realize quantum networks of superconducting qubits with realistic experimental parameters.

quant-ph↗

arXiv2024

Efficient LLM Context Distillation

Large Language Models (LLMs) demonstrate proficiency across diverse tasks but often require targeted adaptations for specific applications. Various methods have been proposed to facilitate this adaptation, including fewshot fine-tuning, in-context learning, and context distillation. This paper specifically investigates context distillation a method that extends the utility of task-specific examples by internalizing them, thus augmenting the example set accessible for model inference. We conduct a comparative analysis of context distillation with in-context learning (ICL) and few-shot fine-tuning (FT), aiming to ascertain the efficacy of context distillation in adapting models using minimal in-context examples. Employing matched datasets from Mobach, our experiments leverage OPT models of various sizes. The results indicate that context distillation effectively adapts models, with student models attaining comparable in-domain and out-of-domain accuracies to in-context learning. Although context distillation surpasses ICL in out-of-domain generalization, it does not achieve the performance levels of FT. However, the reduced dataset size and computational demands position context distillation as a viable alternative, especially for smaller datasets. Overall, this study presents context distillation as an efficient and potent method for customizing LLMs to specific tasks.

cs.LG↗

arXiv2024

Curve attractors for marked rational maps

A Thurston map $f\colon (S^2, A) \to (S^2, A)$ with marking set $A$ induces a pullback relation on isotopy classes of Jordan curves in $(S^2, A)$. If every curve lands in a finite list of possible curve classes after iterating this pullback relation, then the pair $(f,A)$ is said to have a finite global curve attractor. It is conjectured by Pilgrim that all rational Thurston maps that are not flexible Lattès maps have a finite global curve attractor. We present partial progress on this problem. Specifically, we prove that if $A$ has four points and the postcritical set (which is a subset of $A$) has two or three points, then $(f,A)$ has a finite global curve attractor. We also discuss extensions of the main result to certain special cases where $f$ has four postcritical points and $A=P_f$. Additionally, we speculate on how some of these ideas might be used in the more general case.

math.DS↗

arXiv2023

Enhanced Sampling with Machine Learning: A Review

Molecular dynamics (MD) enables the study of physical systems with excellent spatiotemporal resolution but suffers from severe time-scale limitations. To address this, enhanced sampling methods have been developed to improve exploration of configurational space. However, implementing these is challenging and requires domain expertise. In recent years, integration of machine learning (ML) techniques in different domains has shown promise, prompting their adoption in enhanced sampling as well. Although ML is often employed in various fields primarily due to its data-driven nature, its integration with enhanced sampling is more natural with many common underlying synergies. This review explores the merging of ML and enhanced MD by presenting different shared viewpoints. It offers a comprehensive overview of this rapidly evolving field, which can be difficult to stay updated on. We highlight successful strategies like dimensionality reduction, reinforcement learning, and flow-based methods. Finally, we discuss open problems at the exciting ML-enhanced MD interface.

cond-mat.stat-mech↗

arXiv2021

SGOOP-d: Estimating kinetic distances and reaction coordinate dimensionality for rare event systems from biased/unbiased simulations

Understanding kinetics including reaction pathways and associated transition rates is an important yet difficult problem in numerous chemical and biological systems especially in situations with multiple competing pathways. When these high-dimensional systems are projected on low-dimensional coordinates, which are often needed for enhanced sampling or for interpretation of simulations and experiments, one can end up losing the kinetic connectivity of the underlying high-dimensional landscape. Thus in the low-dimensional projection metastable states might appear closer or further than they actually are. To deal with this issue, in this work we develop a formalism that learns a multi-dimensional yet minimally complex reaction coordinate (RC) for generic high-dimensional systems. When projected along this RC, all possible kinetically relevant pathways can be demarcated and the true high-dimensional connectivity is maintained. One of the defining attributes of our method lies in that it can work on long unbiased simulations as well as biased simulations often needed for rare event systems. We demonstrate the utility of the method by studying a range of model systems including conformational transitions in a small peptide Ace-Ala$_3$-Nme, where we show how two-dimensional and three-dimensional reaction coordinate found by our previously published spectral gap optimization method "SGOOP" [P. Tiwary and B. J. Berne, Proc. Natl. Acad. Sci. 113, 2839 (2016)] can capture the kinetics for 23 and all 28 out of the 28 dominant state-to-state transitions respectively.

physics.comp-ph↗

arXiv2019

Reaction coordinates and rate constants for liquid droplet nucleation: quantifying the interplay between driving force and memory

In this work we revisit the classic problem of homogeneous nucleation of a liquid droplet in a supersaturated vapor phase. We consider this at different extents of the driving force, which here is the extent of supersaturation, and calculate a reaction coordinate (RC) for nucleation as the driving force is varied. The RC is constructed as a linear combination of three order parameters, where one accounts for the number of liquid-like atoms, and the other two for local density fluctuations. The RC is calculated from all-atom biased and unbiased molecular dynamics (MD) simulations using the spectral gap optimization approach "SGOOP" [P. Tiwary and B. J. Berne, Proc. Natl. Acad. Sci. U. S. A. 113, 2839 (2016)]. Our key finding is that as the supersaturation decreases, the RC ceases to simply be the number of liquid-like atoms, and instead it becomes important to explicitly consider local density fluctuations that correlate with shape and density variations in the nucleus. All three order parameters are found to have similar barriers in their respective potentials of mean force, however, as the supersaturation decreases the density fluctuations decorrelate slower and thus carry longer memory. Thus at lower supersaturations density fluctuations are non-Markovian and can not be simply ignored from the RC by virtue of being noise. Finally, we use this optimized RC to calculate nucleation rates in the infrequent metadynamics framework, and show it leads to more accurate estimate of the nucleation rate with four orders of magnitude acceleration relative to unbiased MD.

physics.comp-ph↗