Showing 1–39 of 39 results for author: Morse, K

MedHELM: Holistic Evaluation of Large Language Models for Medical Tasks

Authors: Suhana Bedi, Hejie Cui, Miguel Fuentes, Alyssa Unell, Michael Wornow, Juan M. Banda, Nikesh Kotecha, Timothy Keyes, Yifan Mai, Mert Oez, Hao Qiu, Shrey Jain, Leonardo Schettini, Mehr Kashyap, Jason Alan Fries, Akshay Swaminathan, Philip Chung, Fateme Nateghi, Asad Aali, Ashwin Nayak, Shivam Vedak, Sneha S. Jain, Birju Patel, Oluseyi Fayanju, Shreya Shah , et al. (56 additional authors not shown)

Abstract: While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. We introduce MedHELM, an extensible evaluation framework for assessing LLM performance for medical tasks with three key contributions. First, a clinician-validated taxonomy spanning 5 categories, 22 subcatego… ▽ More While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. We introduce MedHELM, an extensible evaluation framework for assessing LLM performance for medical tasks with three key contributions. First, a clinician-validated taxonomy spanning 5 categories, 22 subcategories, and 121 tasks developed with 29 clinicians. Second, a comprehensive benchmark suite comprising 35 benchmarks (17 existing, 18 newly formulated) providing complete coverage of all categories and subcategories in the taxonomy. Third, a systematic comparison of LLMs with improved evaluation methods (using an LLM-jury) and a cost-performance analysis. Evaluation of 9 frontier LLMs, using the 35 benchmarks, revealed significant performance variation. Advanced reasoning models (DeepSeek R1: 66% win-rate; o3-mini: 64% win-rate) demonstrated superior performance, though Claude 3.5 Sonnet achieved comparable results at 40% lower estimated computational cost. On a normalized accuracy scale (0-1), most models performed strongly in Clinical Note Generation (0.73-0.85) and Patient Communication & Education (0.78-0.83), moderately in Medical Research Assistance (0.65-0.75), and generally lower in Clinical Decision Support (0.56-0.72) and Administration & Workflow (0.53-0.63). Our LLM-jury evaluation method achieved good agreement with clinician ratings (ICC = 0.47), surpassing both average clinician-clinician agreement (ICC = 0.43) and automated baselines including ROUGE-L (0.36) and BERTScore-F1 (0.44). Claude 3.5 Sonnet achieved comparable performance to top models at lower estimated cost. These findings highlight the importance of real-world, task-specific evaluation for medical use of LLMs and provides an open source framework to enable this. △ Less

Submitted 2 June, 2025; v1 submitted 26 May, 2025; originally announced May 2025.

Jamming the Random Lorentz Gas: Configurational Entropy, Crunching Geometry, and Critical Universality

Authors: Giampaolo Folena, Patrick Charbonneau, Peter K. Morse, Rafael Díaz Hernández Rojas, Federico Ricci-Tersenghi

Abstract: Deterministic optimization algorithms unequivocally partition a complex energy landscape in inherent structures (ISs) and their respective basins of attraction. But can these basins be defined solely through geometric principles? This question is particularly important when examining the jamming of hard spheres. We here consider this issue by proposing a geometric class of gradient descent-like al… ▽ More Deterministic optimization algorithms unequivocally partition a complex energy landscape in inherent structures (ISs) and their respective basins of attraction. But can these basins be defined solely through geometric principles? This question is particularly important when examining the jamming of hard spheres. We here consider this issue by proposing a geometric class of gradient descent-like algorithms, which we use to study a system in the hard-sphere universality class, the random Lorentz gas. The statistics of the resulting IS is found to be strictly inherited from those of Poisson-Voronoi tessellations. The landscape roughness is further found to give rise to a hierarchical organization of ISs which various algorithms explore differently. In particular, greedy and reluctant schemes tend to favor ISs of markedly different densities. The resulting ISs nevertheless robustly exhibit a universal force distribution, thus confirming the geometric nature of the jamming universality class. △ Less

Submitted 8 October, 2024; originally announced October 2024.

Comments: 16 pages, 11 figures

Simple fluctuations in simple glass formers

Authors: Corentin C. L. Laudicina, Patrick Charbonneau, Yi Hu, Liesbeth M. C. Janssen, Peter K. Morse, Ilian Pihlajamaa, Grzegorz Szamel

Abstract: Critical single-particle fluctuations associated with particle displacements are inherent to simple glass-forming liquids in the limit of large dimensions and leave a pseudo-critical trace across all finite dimensions. This characteristic could serve as a crucial test for distinguishing between theories of glass formation. We here examine these critical fluctuations, as captured by the well-establ… ▽ More Critical single-particle fluctuations associated with particle displacements are inherent to simple glass-forming liquids in the limit of large dimensions and leave a pseudo-critical trace across all finite dimensions. This characteristic could serve as a crucial test for distinguishing between theories of glass formation. We here examine these critical fluctuations, as captured by the well-established non-Gaussian parameter, within both mode-coupling theory (MCT) and dynamical mean-field theory (DMFT) across dimensions for hard sphere liquids and for the minimally structured Mari--Kurchan model. We establish general scaling laws relevant to any liquid dynamics theory in large dimensions and show that the dimensional scalings predicted by MCT are inconsistent with those from DMFT. Simulation results for hard spheres in moderately high dimensions align with the DMFT scenario, reinforcing the relevance of mean-field theory for capturing glass physics in finite dimensions. We identify potential adjustments to MCT to account for certain mean-field physics. Our findings also highlight that local structure and spatial dimensionality can affect single-particle critical fluctuations in non-trivial ways. △ Less

Submitted 13 August, 2024; originally announced August 2024.

Comments: 42 pages, 9 figures

Ordered and disordered stealthy hyperuniform point patterns across spatial dimensions

Authors: Peter K. Morse, Paul J. Steinhardt, Salvatore Torquato

Abstract: In previous work [Phys. Rev. X 5, 021020 (2015)], it was shown that stealthy hyperuniform systems can be regarded as hard spheres in Fourier-space in the sense that the the structure factor is exactly zero in a spherical region around the origin in analogy with the pair-correlation function of real-space hard spheres. In this work, we exploit this correspondence to confirm that the densest Fourier… ▽ More In previous work [Phys. Rev. X 5, 021020 (2015)], it was shown that stealthy hyperuniform systems can be regarded as hard spheres in Fourier-space in the sense that the the structure factor is exactly zero in a spherical region around the origin in analogy with the pair-correlation function of real-space hard spheres. In this work, we exploit this correspondence to confirm that the densest Fourier-space hard-sphere system is that of a Bravais lattice. This is in contrast to real-space hard-spheres, whose densest configuration is conjectured to be disordered. We also extend the virial series previously suggested for disordered stealthy hyperuniform systems to higher dimensions in order to predict spatial decorrelation as function of dimension. This prediction is then borne out by numerical simulations of disordered stealthy hyperuniform ground states in dimensions $d=2$-$8$. △ Less

Submitted 28 April, 2024; v1 submitted 25 April, 2024; originally announced April 2024.

Comments: 9 pages; 6 figures

Masked Autoencoders for Microscopy are Scalable Learners of Cellular Biology

Authors: Oren Kraus, Kian Kenyon-Dean, Saber Saberian, Maryam Fallah, Peter McLean, Jess Leung, Vasudev Sharma, Ayla Khan, Jia Balakrishnan, Safiye Celik, Dominique Beaini, Maciej Sypetkowski, Chi Vicky Cheng, Kristen Morse, Maureen Makes, Ben Mabey, Berton Earnshaw

Abstract: Featurizing microscopy images for use in biological research remains a significant challenge, especially for large-scale experiments spanning millions of images. This work explores the scaling properties of weakly supervised classifiers and self-supervised masked autoencoders (MAEs) when training with increasingly larger model backbones and microscopy datasets. Our results show that ViT-based MAEs… ▽ More Featurizing microscopy images for use in biological research remains a significant challenge, especially for large-scale experiments spanning millions of images. This work explores the scaling properties of weakly supervised classifiers and self-supervised masked autoencoders (MAEs) when training with increasingly larger model backbones and microscopy datasets. Our results show that ViT-based MAEs outperform weakly supervised classifiers on a variety of tasks, achieving as much as a 11.5% relative improvement when recalling known biological relationships curated from public databases. Additionally, we develop a new channel-agnostic MAE architecture (CA-MAE) that allows for inputting images of different numbers and orders of channels at inference time. We demonstrate that CA-MAEs effectively generalize by inferring and evaluating on a microscopy image dataset (JUMP-CP) generated under different experimental conditions with a different channel structure than our pretraining data (RPI-93M). Our findings motivate continued research into scaling self-supervised learning on microscopy data in order to create powerful foundation models of cellular biology that have the potential to catalyze advancements in drug discovery and beyond. △ Less

Submitted 15 April, 2024; originally announced April 2024.

Comments: CVPR 2024 Highlight. arXiv admin note: text overlap with arXiv:2309.16064

Journal ref: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Memory formation in dense persistent active matter

Authors: Elisabeth Agoritsas, Peter K. Morse

Abstract: Protocol-dependent states in structural glasses can encode a disordered, yet retrievable memory. While training such materials is typically done via a global drive, such as external shear, in dense active matter the driving is instead local and spatio-temporally correlated. Here we focus on the impact of such spatial correlation on memory formation. We investigate the mechanical response of a dens… ▽ More Protocol-dependent states in structural glasses can encode a disordered, yet retrievable memory. While training such materials is typically done via a global drive, such as external shear, in dense active matter the driving is instead local and spatio-temporally correlated. Here we focus on the impact of such spatial correlation on memory formation. We investigate the mechanical response of a dense amorphous packing of athermal particles, subject to an oscillatory quasistatic driving with a tunable spatial correlation, akin to the instantaneous driving pattern in active matter. We find that the capacity to encode memory can be rendered comparable upon a proper rescaling on the spatial correlation, whereas the efficiency in memory formation increases with motion cooperativity. △ Less

Submitted 4 April, 2024; v1 submitted 18 March, 2024; originally announced March 2024.

Standing on FURM ground -- A framework for evaluating Fair, Useful, and Reliable AI Models in healthcare systems

Authors: Alison Callahan, Duncan McElfresh, Juan M. Banda, Gabrielle Bunney, Danton Char, Jonathan Chen, Conor K. Corbin, Debadutta Dash, Norman L. Downing, Sneha S. Jain, Nikesh Kotecha, Jonathan Masterson, Michelle M. Mello, Keith Morse, Srikar Nallan, Abby Pandya, Anurang Revri, Aditya Sharma, Christopher Sharp, Rahul Thapa, Michael Wornow, Alaa Youssef, Michael A. Pfeffer, Nigam H. Shah

Abstract: The impact of using artificial intelligence (AI) to guide patient care or operational processes is an interplay of the AI model's output, the decision-making protocol based on that output, and the capacity of the stakeholders involved to take the necessary subsequent action. Estimating the effects of this interplay before deployment, and studying it in real time afterwards, are essential to bridge… ▽ More The impact of using artificial intelligence (AI) to guide patient care or operational processes is an interplay of the AI model's output, the decision-making protocol based on that output, and the capacity of the stakeholders involved to take the necessary subsequent action. Estimating the effects of this interplay before deployment, and studying it in real time afterwards, are essential to bridge the chasm between AI model development and achievable benefit. To accomplish this, the Data Science team at Stanford Health Care has developed a Testing and Evaluation (T&E) mechanism to identify fair, useful and reliable AI models (FURM) by conducting an ethical review to identify potential value mismatches, simulations to estimate usefulness, financial projections to assess sustainability, as well as analyses to determine IT feasibility, design a deployment strategy, and recommend a prospective monitoring and evaluation plan. We report on FURM assessments done to evaluate six AI guided solutions for potential adoption, spanning clinical and operational settings, each with the potential to impact from several dozen to tens of thousands of patients each year. We describe the assessment process, summarize the six assessments, and share our framework to enable others to conduct similar assessments. Of the six solutions we assessed, two have moved into a planning and implementation phase. Our novel contributions - usefulness estimates by simulation, financial projections to quantify sustainability, and a process to do ethical assessments - as well as their underlying methods and open source tools, are available for other healthcare systems to conduct actionable evaluations of candidate AI solutions. △ Less

Submitted 14 March, 2024; v1 submitted 26 February, 2024; originally announced March 2024.

Critical slowing down in thermal soft-sphere glasses via energy minimization

Authors: Kevin A. Interiano-Alberto, Peter K. Morse, Robert S. Hoy

Abstract: Using hybrid molecular dynamics/SWAP Monte Carlo (MD/SMC) simulations, we show that the terminal relaxation times $τ$ for FIRE energy minimization of soft-sphere glasses exhibit thermal onset as samples become increasingly well-equilibrated. Although $τ(φ)$ can decrease by orders of magnitude as equilibration proceeds and the jamming density $φ_{\rm J}$ increases via thermal onset, it always scale… ▽ More Using hybrid molecular dynamics/SWAP Monte Carlo (MD/SMC) simulations, we show that the terminal relaxation times $τ$ for FIRE energy minimization of soft-sphere glasses exhibit thermal onset as samples become increasingly well-equilibrated. Although $τ(φ)$ can decrease by orders of magnitude as equilibration proceeds and the jamming density $φ_{\rm J}$ increases via thermal onset, it always scales as $τ(φ) \sim (φ_{\rm J} - φ)^{-2} \sim [Z_{\rm iso} - Z_{\rm ms}(τ)]^{-2}$, where $φ_{\rm J}$ is the jamming density and $Z_{\rm ms}(τ)$ is the average coordination number of particles satisfying a minimal local mechanical stability criterion ($Z \geq d+1$) at the top of the final potential-energy-landscape (PEL) sub-basin the system encounters. This scaling allows us to collapse $τ$ datasets that look very different when plotted as a function of $φ$, and to address a closely related question: how does the character of the PEL basins that dense thermal glasses most typically occupy evolve as the glasses age at constant $φ$ and $T$? △ Less

Submitted 15 April, 2024; v1 submitted 20 December, 2023; originally announced December 2023.

Comments: resubmitted in response to referee comments

Dynamics and fluctuations of minimally-structured glass formers

Authors: Patrick Charbonneau, Yi Hu, Peter K. Morse

Abstract: The mean-field theory (MFT) of simple glasses, which is exact in the limit of infinite spatial dimensions, $d\rightarrow\infty$, offers theoretical insight as well as quantitative predictions about certain features of $d=3$ systems. In order to more systematically relate the behavior of physical systems to MFT, however, various finite-$d$ corrections need to be accounted for. Although some efforts… ▽ More The mean-field theory (MFT) of simple glasses, which is exact in the limit of infinite spatial dimensions, $d\rightarrow\infty$, offers theoretical insight as well as quantitative predictions about certain features of $d=3$ systems. In order to more systematically relate the behavior of physical systems to MFT, however, various finite-$d$ corrections need to be accounted for. Although some efforts along this direction have already been undertaken, theoretical and technical challenges hinder progress. A general approach to sidestep many of these difficulties consists of simulating minimally-structured models whose behavior smoothly converges to that described by the MFT as $d$ increases, so as to permit a dimensional extrapolation. Using this approach, we here extract the small fluctuations around the dynamical MFT captured by a standard liquid-state observable the non-Gaussian parameter $α_2$. The results provide insight into the physical origin of these fluctuations as well as a quantitative reference with which to compare observations in more realistic glass formers. △ Less

Submitted 12 December, 2023; originally announced December 2023.

Comments: 10 pages, 7 figures

Journal ref: Phys. Rev. E 109, 054905 (2024)

A Multi-Center Study on the Adaptability of a Shared Foundation Model for Electronic Health Records

Authors: Lin Lawrence Guo, Jason Fries, Ethan Steinberg, Scott Lanyon Fleming, Keith Morse, Catherine Aftandilian, Jose Posada, Nigam Shah, Lillian Sung

Abstract: Foundation models hold promise for transforming AI in healthcare by providing modular components that are easily adaptable to downstream healthcare tasks, making AI development more scalable and cost-effective. Structured EHR foundation models, trained on coded medical records from millions of patients, demonstrated benefits including increased performance with fewer training labels, and improved… ▽ More Foundation models hold promise for transforming AI in healthcare by providing modular components that are easily adaptable to downstream healthcare tasks, making AI development more scalable and cost-effective. Structured EHR foundation models, trained on coded medical records from millions of patients, demonstrated benefits including increased performance with fewer training labels, and improved robustness to distribution shifts. However, questions remain on the feasibility of sharing these models across different hospitals and their performance for local task adaptation. This multi-center study examined the adaptability of a recently released structured EHR foundation model ($FM_{SM}$), trained on longitudinal medical record data from 2.57M Stanford Medicine patients. Experiments were conducted using EHR data at The Hospital for Sick Children and MIMIC-IV. We assessed both adaptability via continued pretraining on local data, and task adaptability compared to baselines of training models from scratch at each site, including a local foundation model. We evaluated the performance of these models on 8 clinical prediction tasks. In both datasets, adapting the off-the-shelf $FM_{SM}$ matched the performance of GBM models locally trained on all data while providing a 13% improvement in settings with few task-specific training labels. With continued pretraining on local data, label efficiency substantially improved, such that $FM_{SM}$ required fewer than 1% of training examples to match the fully trained GBM's performance. Continued pretraining was also 60 to 90% more sample-efficient than training local foundation models from scratch. Our findings show that adapting shared EHR foundation models across hospitals provides improved prediction performance at less cost, underscoring the utility of base foundation models as modular components to streamline the development of healthcare AI. △ Less

Submitted 22 April, 2024; v1 submitted 19 November, 2023; originally announced November 2023.

Comments: 46 pages, 5 figures, 3 tables, 14 appendices

Local stability of spheres via the convex hull and the radical Voronoi diagram

Authors: Peter K. Morse, Eric Corwin

Abstract: Jamming is an emergent phenomenon wherein the local stability of individual particles percolates to form a globally rigid structure. However, the onset of rigidity does not imply that every particle becomes rigid, and indeed some remain locally unstable. These particles, if they become unmoored from their neighbors, are called \textit{rattlers}, and their identification is critical to understandin… ▽ More Jamming is an emergent phenomenon wherein the local stability of individual particles percolates to form a globally rigid structure. However, the onset of rigidity does not imply that every particle becomes rigid, and indeed some remain locally unstable. These particles, if they become unmoored from their neighbors, are called \textit{rattlers}, and their identification is critical to understanding the rigid backbone of a packing, as these particles cannot bear stress. The accurate identification of rattlers, however, can be a time-consuming process, and the currently accepted method lacks a simple geometric interpretation. In this manuscript, we propose two simpler classifications of rattlers based on the convex hull of contacting neighbors and the maximum inscribed sphere of the radical Voronoi cell, each of which provides geometric insight into the source of their instability. Furthermore, the convex hull formulation can be generalized to explore stability in hyperstatic soft sphere packings, spring networks, non-spherical packings, and mean-field non-central-force potentials. △ Less

Submitted 28 September, 2023; originally announced September 2023.

Comments: 9 pages, 9 figures

Masked Autoencoders are Scalable Learners of Cellular Morphology

Authors: Oren Kraus, Kian Kenyon-Dean, Saber Saberian, Maryam Fallah, Peter McLean, Jess Leung, Vasudev Sharma, Ayla Khan, Jia Balakrishnan, Safiye Celik, Maciej Sypetkowski, Chi Vicky Cheng, Kristen Morse, Maureen Makes, Ben Mabey, Berton Earnshaw

Abstract: Inferring biological relationships from cellular phenotypes in high-content microscopy screens provides significant opportunity and challenge in biological research. Prior results have shown that deep vision models can capture biological signal better than hand-crafted features. This work explores how self-supervised deep learning approaches scale when training larger models on larger microscopy d… ▽ More Inferring biological relationships from cellular phenotypes in high-content microscopy screens provides significant opportunity and challenge in biological research. Prior results have shown that deep vision models can capture biological signal better than hand-crafted features. This work explores how self-supervised deep learning approaches scale when training larger models on larger microscopy datasets. Our results show that both CNN- and ViT-based masked autoencoders significantly outperform weakly supervised baselines. At the high-end of our scale, a ViT-L/8 trained on over 3.5-billion unique crops sampled from 93-million microscopy images achieves relative improvements as high as 28% over our best weakly supervised baseline at inferring known biological relationships curated from public databases. Relevant code and select models released with this work can be found at: https://github.com/recursionpharma/maes_microscopy. △ Less

Submitted 27 November, 2023; v1 submitted 27 September, 2023; originally announced September 2023.

Comments: Spotlight at NeurIPS 2023 Generative AI and Biology (GenBio) Workshop

MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records

Authors: Scott L. Fleming, Alejandro Lozano, William J. Haberkorn, Jenelle A. Jindal, Eduardo P. Reis, Rahul Thapa, Louis Blankemeier, Julian Z. Genkins, Ethan Steinberg, Ashwin Nayak, Birju S. Patel, Chia-Chun Chiang, Alison Callahan, Zepeng Huo, Sergios Gatidis, Scott J. Adams, Oluseyi Fayanju, Shreya J. Shah, Thomas Savage, Ethan Goh, Akshay S. Chaudhari, Nima Aghaeepour, Christopher Sharp, Michael A. Pfeffer, Percy Liang , et al. (5 additional authors not shown)

Abstract: The ability of large language models (LLMs) to follow natural language instructions with human-level fluency suggests many opportunities in healthcare to reduce administrative burden and improve quality of care. However, evaluating LLMs on realistic text generation tasks for healthcare remains challenging. Existing question answering datasets for electronic health record (EHR) data fail to capture… ▽ More The ability of large language models (LLMs) to follow natural language instructions with human-level fluency suggests many opportunities in healthcare to reduce administrative burden and improve quality of care. However, evaluating LLMs on realistic text generation tasks for healthcare remains challenging. Existing question answering datasets for electronic health record (EHR) data fail to capture the complexity of information needs and documentation burdens experienced by clinicians. To address these challenges, we introduce MedAlign, a benchmark dataset of 983 natural language instructions for EHR data. MedAlign is curated by 15 clinicians (7 specialities), includes clinician-written reference responses for 303 instructions, and provides 276 longitudinal EHRs for grounding instruction-response pairs. We used MedAlign to evaluate 6 general domain LLMs, having clinicians rank the accuracy and quality of each LLM response. We found high error rates, ranging from 35% (GPT-4) to 68% (MPT-7B-Instruct), and an 8.3% drop in accuracy moving from 32k to 2k context lengths for GPT-4. Finally, we report correlations between clinician rankings and automated natural language generation metrics as a way to rank LLMs without human review. We make MedAlign available under a research data use agreement to enable LLM evaluations on tasks aligned with clinician needs and preferences. △ Less

Submitted 24 December, 2023; v1 submitted 27 August, 2023; originally announced August 2023.

Jamming, relaxation, and memory in a structureless glass former

Authors: Patrick Charbonneau, Peter K. Morse

Abstract: Real structural glasses form through various out-of-equilibrium processes, including temperature quenches, rapid compression, shear, and aging. Each of these processes should be formally understandable within the recently formulated dynamical mean-field theory of glasses, but many of the numerical tools needed to solve the relevant equations for sufficiently long timescales do not yet exist. Numer… ▽ More Real structural glasses form through various out-of-equilibrium processes, including temperature quenches, rapid compression, shear, and aging. Each of these processes should be formally understandable within the recently formulated dynamical mean-field theory of glasses, but many of the numerical tools needed to solve the relevant equations for sufficiently long timescales do not yet exist. Numerical simulations of structureless (and therefore mean-field-like) model glass formers can nevertheless aid searching for and understanding such solutions, thanks to their ability to disentangle structural from dimensional effects. We here study the infinite-range Mari-Kurchan model under simple non-equilibrium processes and compare the results with those from the random Lorentz gas [J. Phys. A: Math. Theor. 55 334001, (2022)], which are both mean-field-like and become formally equivalent in the limit of infinite spatial dimensions. Of particular interest are jamming from crunching and under instantaneous temperature quenches. The study allows for an algorithmic understanding of the jamming density and of its approach to the infinite-dimensional limit. The results provide important insight into the eventual solution of the dynamical mean-field theory, including onsets and anomalous relaxation, as well as into the various algorithmic schemes for jamming. △ Less

Submitted 21 April, 2023; originally announced April 2023.

Comments: 13 pages, 6 figures

Journal ref: Phys. Rev. E 108, 054102 (2023)

Generating large disordered stealthy hyperuniform systems with ultra-high accuracy to determine their physical properties

Authors: Peter K. Morse, Jaeuk Kim, Paul J. Steinhardt, Salvatore Torquato

Abstract: Hyperuniform many-particle systems are characterized by a structure factor $S({\mathbf{k}})$ that is precisely zero as $|\mathbf{k}|\rightarrow0$; and stealthy hyperuniform systems have $S({\mathbf{k}})=0$ for the finite range $0 < |{\mathbf{k}}| \le K$, called the "exclusion region." Through a process of collective-coordinate optimization, energy-minimizing disordered stealthy hyperuniform system… ▽ More Hyperuniform many-particle systems are characterized by a structure factor $S({\mathbf{k}})$ that is precisely zero as $|\mathbf{k}|\rightarrow0$; and stealthy hyperuniform systems have $S({\mathbf{k}})=0$ for the finite range $0 < |{\mathbf{k}}| \le K$, called the "exclusion region." Through a process of collective-coordinate optimization, energy-minimizing disordered stealthy hyperuniform systems of moderate size have been made to high accuracy, and their novel physical properties have shown great promise. However, minimizing $S(\mathbf{k})$ in the exclusion region is computationally intensive as the system size becomes large. In this Letter, we present an improved methodology to generate such states using double-double precision calculations on GPUs that reduces the deviations from zero within the exclusion region by a factor of approximately $10^{30}$ for systems sizes more than an order of magnitude larger. We further show that this ultra-high accuracy is required to draw conclusions about their corresponding characteristics, such as the nature of the associated energy landscape and the presence or absence of Anderson localization, which might be masked, even when deviations are relatively small. △ Less

Submitted 18 April, 2023; originally announced April 2023.

Comments: 7 pages, 3 figures

Instability in clinical risk stratification models using deep learning

Authors: Daniel Lopez-Martinez, Alex Yakubovich, Martin Seneviratne, Adam D. Lelkes, Akshit Tyagi, Jonas Kemp, Ethan Steinberg, N. Lance Downing, Ron C. Li, Keith E. Morse, Nigam H. Shah, Ming-Jun Chen

Abstract: While it has been well known in the ML community that deep learning models suffer from instability, the consequences for healthcare deployments are under characterised. We study the stability of different model architectures trained on electronic health records, using a set of outpatient prediction tasks as a case study. We show that repeated training runs of the same deep learning model on the sa… ▽ More While it has been well known in the ML community that deep learning models suffer from instability, the consequences for healthcare deployments are under characterised. We study the stability of different model architectures trained on electronic health records, using a set of outpatient prediction tasks as a case study. We show that repeated training runs of the same deep learning model on the same training data can result in significantly different outcomes at a patient level even though global performance metrics remain stable. We propose two stability metrics for measuring the effect of randomness of model training, as well as mitigation strategies for improving model stability. △ Less

Submitted 19 November, 2022; originally announced November 2022.

Comments: Accepted for publication in Machine Learning for Health (ML4H) 2022

Comment on "Explicit Analytical Solution for Random Close Packing in $d = 2$ and $d = 3$"

Authors: Patrick Charbonneau, Peter K. Morse

Abstract: A recent letter titled "Explicit Analytical Solution for Random Close Packing in d=2 and d=3" published in Physical Review Letters proposes a first-principle computation of the random close packing (RCP) density in spatial dimensions d=2 and d=3. This problem has a long history of such proposals, but none capture the full picture. This paper, in particular, once generalized to all d fails to descr… ▽ More A recent letter titled "Explicit Analytical Solution for Random Close Packing in d=2 and d=3" published in Physical Review Letters proposes a first-principle computation of the random close packing (RCP) density in spatial dimensions d=2 and d=3. This problem has a long history of such proposals, but none capture the full picture. This paper, in particular, once generalized to all d fails to describe the known behavior of jammed systems in d>4, thus suggesting that the low-dimensional agreement is largely fortuitous. △ Less

Submitted 19 January, 2022; originally announced January 2022.

Comments: 2 pages, 1 figure

The dimensional evolution of structure and dynamics in hard sphere liquids

Authors: Patrick Charbonneau, Yi Hu, Joyjit Kundu, Peter K. Morse

Abstract: The formulation of the mean-field, infinite-dimensional solution of hard sphere glasses is a significant milestone for theoretical physics. How relevant this description might be for understanding low-dimensional glass-forming liquids, however, remains unclear. These liquids indeed exhibit a complex interplay between structure and dynamics, and the importance of this interplay might only slowly di… ▽ More The formulation of the mean-field, infinite-dimensional solution of hard sphere glasses is a significant milestone for theoretical physics. How relevant this description might be for understanding low-dimensional glass-forming liquids, however, remains unclear. These liquids indeed exhibit a complex interplay between structure and dynamics, and the importance of this interplay might only slowly diminish as dimension $d$ increases. A careful numerical assessment of the matter has long been hindered by the exponential increase of computational costs with $d$. By revisiting a once common simulation technique involving the use of periodic boundary conditions modeled on $D_d$ lattices, we here partly sidestep this difficulty, thus allowing the study of hard sphere liquids up to $d=13$. Parallel efforts by Mangeat and Zamponi [Phys. Rev. E 93, 012609 (2016)] have expanded the mean-field description of glasses to finite $d$ by leveraging standard liquid-state theory, and thus help bridge the gap from the other direction. The relatively smooth evolution of both structure and dynamics across the $d$ gap allows us to relate the two approaches, and to identify some of the missing features that a finite-$d$ theory of glasses might hope to include to achieve near quantitative agreement. △ Less

Submitted 26 November, 2021; originally announced November 2021.

Comments: 11 pages, 8 figures

Journal ref: J. Chem. Phys. 156, 134502 (2022)

Three simple scenarios for high-dimensional sphere packings

Authors: Patrick Charbonneau, Peter K. Morse, Will Perkins, Francesco Zamponi

Abstract: Based on results from the physics and mathematics literature which suggest a series of clearly defined conjectures, we formulate three simple scenarios for the fate of hard sphere crystallization in high dimension: (A) crystallization is impeded and the glass phase constitutes the densest packing, (B) crystallization from the liquid is possible, but takes place much beyond the dynamical glass tran… ▽ More Based on results from the physics and mathematics literature which suggest a series of clearly defined conjectures, we formulate three simple scenarios for the fate of hard sphere crystallization in high dimension: (A) crystallization is impeded and the glass phase constitutes the densest packing, (B) crystallization from the liquid is possible, but takes place much beyond the dynamical glass transition and is thus dynamically implausible, or (C) crystallization is possible and takes place before (or just after) dynamical arrest, thus making it plausibly accessible from the liquid state. In order to assess the underlying conjectures and thus obtain insight into which scenario is most likely to be realized, we investigate the densest sphere packings in dimension $d=3$-$10$ using cell-cluster expansions as well as numerical simulations. These resulting estimates of the crystal entropy near close-packing tend to support scenario C. We additionally confirm that the crystal equation of state is dominated by the free volume expansion and that a meaningful polynomial correction can be formulated. △ Less

Submitted 28 December, 2021; v1 submitted 7 September, 2021; originally announced September 2021.

Comments: 17 pages, 9 figures. Edits coincide with the published version

Journal ref: Phys. Rev. E 104, 064612 (2021)

Thermodynamic stability of hard sphere crystals in dimensions 3 through 10

Authors: Patrick Charbonneau, Caitlin M. Gish, Robert S. Hoy, Peter K. Morse

Abstract: Although much is known about the metastable liquid branch of hard spheres--from low dimension $d$ up to ${d\to\infty}$--its crystal counterpart remains largely unexplored for $d>3$. In particular, it is unclear whether the crystal phase is thermodynamically stable in high dimensions and thus whether a mean-field theory of crystals can ever be exact. In order to determine the stability range of har… ▽ More Although much is known about the metastable liquid branch of hard spheres--from low dimension $d$ up to ${d\to\infty}$--its crystal counterpart remains largely unexplored for $d>3$. In particular, it is unclear whether the crystal phase is thermodynamically stable in high dimensions and thus whether a mean-field theory of crystals can ever be exact. In order to determine the stability range of hard sphere crystals, their equation of state is here estimated from numerical simulations, and fluid-crystal coexistence conditions are determined using a generalized Frenkel-Ladd scheme to compute absolute crystal free energies. The results show that the crystal phase is stable at least up to $d=10$, and the dimensional trends suggest that crystal stability likely persists well beyond that point. △ Less

Submitted 22 December, 2021; v1 submitted 29 April, 2021; originally announced April 2021.

Comments: 12 pages, 7 figures. Edits are reflective of the published version and thus include results for d=10

Journal ref: Euro. Phys. J. E 44, 101 (2021)

Optical observation of single spins in silicon

Authors: A. T. K. Kurkjian, D. B. Higginbottom, C. Chartrand, E. R. MacQuarrie, J. R. Klein, N. R. Lee-Hone, J. Stacho, C. Bowness, L. Bergeron, A. DeAbreu, N. A. Brunelle, S. R. Harrigan, J. Kanaganayagam, M. Kazemi, D. W. Marsden, T. S. Richards, L. A. Stott, S. Roorda, K. J. Morse, M. L. W. Thewalt, S. Simmons

Abstract: The global quantum internet will require long-lived, telecommunications band photon-matter interfaces manufactured at scale. Preliminary quantum networks based upon photon-matter interfaces which meet a subset of these demands are encouraging efforts to identify new high-performance alternatives. Silicon is an ideal host for commercial-scale solid-state quantum technologies. It is already an advan… ▽ More The global quantum internet will require long-lived, telecommunications band photon-matter interfaces manufactured at scale. Preliminary quantum networks based upon photon-matter interfaces which meet a subset of these demands are encouraging efforts to identify new high-performance alternatives. Silicon is an ideal host for commercial-scale solid-state quantum technologies. It is already an advanced platform within the global integrated photonics and microelectronics industries, as well as host to record-setting long-lived spin qubits. Despite the overwhelming potential of the silicon quantum platform, the optical detection of individually addressable photon-spin interfaces in silicon has remained elusive. In this work we produce tens of thousands of individually addressable `$T$ centre' photon-spin qubits in integrated silicon photonic structures, and characterize their spin-dependent telecommunications-band optical transitions. These results unlock immediate opportunities to construct silicon-integrated, telecommunications-band quantum information networks. △ Less

Submitted 12 March, 2021; originally announced March 2021.

Comments: 25 pages, 10 figures

T centres in photonic silicon-on-insulator material

Authors: E. R. MacQuarrie, C. Chartrand, D. B. Higginbottom, K. J. Morse, V. A. Karasyuk, S. Roorda, S. Simmons

Abstract: Global quantum networks will benefit from the reliable creation and control of high-performance solid-state telecom photon-spin interfaces. T radiation damage centres in silicon provide a promising photon-spin interface due to their narrow O-band optical transition near 1326 nm and long-lived electron and nuclear spin lifetimes. To date, these defect centres have only been studied as ensembles in… ▽ More Global quantum networks will benefit from the reliable creation and control of high-performance solid-state telecom photon-spin interfaces. T radiation damage centres in silicon provide a promising photon-spin interface due to their narrow O-band optical transition near 1326 nm and long-lived electron and nuclear spin lifetimes. To date, these defect centres have only been studied as ensembles in bulk silicon. Here, we demonstrate the reliable creation of high concentration T centre ensembles in the 220 nm device layer of silicon-on-insulator (SOI) wafers by ion implantation and subsequent annealing. We then develop a method that uses spin-dependent optical transitions to benchmark the characteristic optical spectral diffusion within these T centre ensembles. Using this new technique, we show that with minimal optimization to the fabrication process high densities of implanted T centres localized $\lesssim$100 nm from an interface display ~1 GHz characteristic levels of total spectral diffusion. △ Less

Submitted 5 March, 2021; originally announced March 2021.

Comments: 10 pages, 9 figures

A direct link between active matter and sheared granular systems

Authors: Peter K. Morse, Sudeshna Roy, Elisabeth Agoritsas, Ethan Stanifer, Eric I. Corwin, M. Lisa Manning

Abstract: The similarity in mechanical properties of dense active matter and sheared amorphous solids has been noted in recent years without a rigorous examination of the underlying mechanism. We develop a mean-field model that predicts that their critical behavior should be equivalent in infinite dimensions, up to a rescaling factor that depends on the correlation length of the applied field. We test these… ▽ More The similarity in mechanical properties of dense active matter and sheared amorphous solids has been noted in recent years without a rigorous examination of the underlying mechanism. We develop a mean-field model that predicts that their critical behavior should be equivalent in infinite dimensions, up to a rescaling factor that depends on the correlation length of the applied field. We test these predictions in 2d using a new numerical protocol, termed `athermal quasi-static random displacement', and find that these mean-field predictions are surprisingly accurate in low dimensions. We identify a general class of perturbations that smoothly interpolate between the uncorrelated localized forces that occur in the high-persistence limit of dense active matter, and system-spanning correlated displacements that occur under applied shear. These results suggest a universal framework for predicting flow, deformation, and failure in active and sheared disordered materials. △ Less

Submitted 2 August, 2021; v1 submitted 16 September, 2020; originally announced September 2020.

Characterization of the T center in $^{28}$Si

Authors: L. Bergeron, C. Chartrand, A. T. K. Kurkjian, K. J. Morse, H. Riemann, N. V. Abrosimov, P. Becker, H. -J. Pohl, M. L. W. Thewalt, S. Simmons

Abstract: Silicon is host to two separate leading quantum technology platforms: integrated silicon photonics as well as long-lived spin qubits. There is an ongoing search for the ideal photon-spin interface able to hybridize these two approaches into a single silicon platform offering substantially expanded capabilities. A number of silicon defects are known to have spin-selective optical transitions, altho… ▽ More Silicon is host to two separate leading quantum technology platforms: integrated silicon photonics as well as long-lived spin qubits. There is an ongoing search for the ideal photon-spin interface able to hybridize these two approaches into a single silicon platform offering substantially expanded capabilities. A number of silicon defects are known to have spin-selective optical transitions, although very few of these are known to be in the highly desirable telecommunications bands, and those that do often do not couple strongly to light. Here we characterize the T center in silicon, a highly stable silicon defect which supports a short-lived bound exciton that upon recombination emits light in the telecommunications O-band. In this first study of T centers in $^{28}$Si, we present the temperature dependence of the zero phonon line, report ensemble zero phonon linewidths as narrow as 33(2) MHz, and elucidate the excited state spectrum of the bound exciton. Magneto-photoluminescence, in conjunction with magnetic resonance, is used to observe twelve distinct orientational subsets of the T center, which are independently addressable due to the anisotropic g factor of the bound exciton's hole spin. The T center is thus a promising contender for the hybridization of silicon's two leading quantum technology platforms. △ Less

Submitted 15 June, 2020; originally announced June 2020.

Comments: 13 pages, 10 figures

A silicon-integrated telecom photon-spin interface

Authors: L. Bergeron, C. Chartrand, A. T. K. Kurkjian, K. J. Morse, H. Riemann, N. V. Abrosimov, P. Becker, H. -J. Pohl, M. L. W. Thewalt, S. Simmons

Abstract: Long-distance entanglement distribution is a vital capability for quantum technologies. An outstanding practical milestone towards this aim is the identification of a suitable matter-photon interface which possesses, simultaneously, long coherence lifetimes and efficient telecommunications-band optical access. In this work, alongside its sister publication, we report upon the T center, a silicon d… ▽ More Long-distance entanglement distribution is a vital capability for quantum technologies. An outstanding practical milestone towards this aim is the identification of a suitable matter-photon interface which possesses, simultaneously, long coherence lifetimes and efficient telecommunications-band optical access. In this work, alongside its sister publication, we report upon the T center, a silicon defect with spin-selective optical transitions at 1326 nm in the telecommunications O-band. Here we show that the T center in $^{28}$Si offers electron and nuclear spin lifetimes beyond a millisecond and second respectively, as well as optical lifetimes of 0.94(1) $μ$s and a Debye-Waller factor of 0.23(1). This work represents a significant step towards coherent photonic interconnects between long-lived silicon spins, spin-entangled telecom single-photon emitters, and spin-dependent silicon-integrated photonic nonlinearities for future global quantum technologies. △ Less

Submitted 15 June, 2020; originally announced June 2020.

Comments: 5 pages, 3 figures

Journal ref: PRX Quantum 1, 020301 (2020)

Predicting plasticity in disordered solids from structural indicators

Authors: D. Richard, M. Ozawa, S. Patinet, E. Stanifer, B. Shang, S. A. Ridout, B. Xu, G. Zhang, P. K. Morse, J. -L. Barrat, L. Berthier, M. L. Falk, P. Guan, A. J. Liu, K. Martens, S. Sastry, D. Vandembroucq, E. Lerner, M. L. Manning

Abstract: Amorphous solids lack long-range order. Therefore identifying structural defects -- akin to dislocations in crystalline solids -- that carry plastic flow in these systems remains a daunting challenge. By comparing many different structural indicators in computational models of glasses, under a variety of conditions we carefully assess which of these indicators are able to robustly identify the str… ▽ More Amorphous solids lack long-range order. Therefore identifying structural defects -- akin to dislocations in crystalline solids -- that carry plastic flow in these systems remains a daunting challenge. By comparing many different structural indicators in computational models of glasses, under a variety of conditions we carefully assess which of these indicators are able to robustly identify the structural defects responsible for plastic flow in amorphous solids. We further demonstrate that the density of defects changes as a function of material preparation and strain in a manner that is highly correlated with the macroscopic material response. Our work represents an important step towards predicting how and when an amorphous solid will fail from its microscopic structure. △ Less

Submitted 23 November, 2020; v1 submitted 25 March, 2020; originally announced March 2020.

Journal ref: Phys. Rev. Materials 4, 113609 (2020)

Learning State-Dependent Losses for Inverse Dynamics Learning

Authors: Kristen Morse, Neha Das, Yixin Lin, Austin S. Wang, Akshara Rai, Franziska Meier

Abstract: Being able to quickly adapt to changes in dynamics is paramount in model-based control for object manipulation tasks. In order to influence fast adaptation of the inverse dynamics model's parameters, data efficiency is crucial. Given observed data, a key element to how an optimizer updates model parameters is the loss function. In this work, we propose to apply meta-learning to learn structured, s… ▽ More Being able to quickly adapt to changes in dynamics is paramount in model-based control for object manipulation tasks. In order to influence fast adaptation of the inverse dynamics model's parameters, data efficiency is crucial. Given observed data, a key element to how an optimizer updates model parameters is the loss function. In this work, we propose to apply meta-learning to learn structured, state-dependent loss functions during a meta-training phase. We then replace standard losses with our learned losses during online adaptation tasks. We evaluate our proposed approach on inverse dynamics learning tasks, both in simulation and on real hardware data. In both settings, the structured and state-dependent learned losses improve online adaptation speed, when compared to standard, state-independent loss functions. △ Less

Submitted 14 August, 2020; v1 submitted 10 March, 2020; originally announced March 2020.

Comments: 9 pages, 8 figures, accepted to IROS 2020, * Kristen Morse and Neha Das had equal contribution

Two classes of events in sheared particulate matter

Authors: Peter K. Morse, Sven Wijtmans, Merlijn van Deen, Martin van Hecke, M. Lisa Manning

Abstract: Under shear, a system of particles changes its contact network and becomes unstable as it transitions between mechanically stable states. For hard spheres at zero pressure, contact breaking events necessarily generate an instability, but this is not the case at finite pressure, where we identify two types of contact changes: network events that do not correspond to instabilities and rearrangement… ▽ More Under shear, a system of particles changes its contact network and becomes unstable as it transitions between mechanically stable states. For hard spheres at zero pressure, contact breaking events necessarily generate an instability, but this is not the case at finite pressure, where we identify two types of contact changes: network events that do not correspond to instabilities and rearrangement events that do. The relative fraction of such events is constant as a function of system size, pressure and interaction potential, consistent with our observation that both nonlinearities obey the same finite-size scaling. Thus, the zero-pressure limit of the nonlinear response is highly singular. △ Less

Submitted 23 July, 2019; originally announced July 2019.

Journal ref: Phys. Rev. Research 2, 023179 (2020)

A Mg-pair isoelectronic bound exciton identified by its isotopic fingerprint in $^{28}$Si

Authors: Rohan Abraham, Adam DeAbreu, Kevin Morse, Valentina Shuman, Leonid Portsel, Anatoly Lodygin, Yuri Astrov, Nikolay Abrosimov, Sergey Pavlov, Heinz-Wilhelm Hübers, Stephanie Simmons, Michael Thewalt

Abstract: We use the greatly improved optical linewidths provided by highly enriched $^{28}$Si to study a photoluminescence line near 1017 meV previously observed in the luminescence spectrum of natural Si diffused with Mg, and suggested to result from the recombination of an isoelectronic bound exciton localized at a Mg-pair center. In $^{28}$Si this no-phonon line is found to be comprised of five componen… ▽ More We use the greatly improved optical linewidths provided by highly enriched $^{28}$Si to study a photoluminescence line near 1017 meV previously observed in the luminescence spectrum of natural Si diffused with Mg, and suggested to result from the recombination of an isoelectronic bound exciton localized at a Mg-pair center. In $^{28}$Si this no-phonon line is found to be comprised of five components whose relative intensities closely match the relative abundances of Mg-pairs formed by random combinations of the three stable isotopes of Mg, thus confirming the Mg-pair hypothesis. We further present the results of temperature dependence studies of this center that reveal unusual and as yet unexplained behaviour. △ Less

Submitted 28 September, 2018; originally announced October 2018.

Comments: 5 pages, 4 figures

Journal ref: Phys. Rev. B 98, 205203 (2018)

Characterization of the Si:Se+ spin-photon interface

Authors: Adam DeAbreu, Camille Bowness, Rohan J. S. Abraham, Alzbeta Medvedova, Kevin J. Morse, Helge Riemann, Nikolay V. Abrosimov, Peter Becker, Hans-Joachim Pohl, Michael L. W. Thewalt, Stephanie Simmons

Abstract: Silicon is the most developed electronic and photonic technological platform and hosts some of the highest-performance spin and photonic qubits developed to date. A hybrid quantum technology harnessing an efficient spin-photon interface in silicon would unlock considerable potential by enabling ultra-long-lived photonic memories, distributed quantum networks, microwave to optical photon converters… ▽ More Silicon is the most developed electronic and photonic technological platform and hosts some of the highest-performance spin and photonic qubits developed to date. A hybrid quantum technology harnessing an efficient spin-photon interface in silicon would unlock considerable potential by enabling ultra-long-lived photonic memories, distributed quantum networks, microwave to optical photon converters, and spin-based quantum processors, all linked using integrated silicon photonics. However, the indirect bandgap of silicon makes identification of efficient spin-photon interfaces nontrivial. Here we build upon the recent identification of chalcogen donors as a promising spin-photon interface in silicon. We determined that the spin-dependent optical degree of freedom has a transition dipole moment stronger than previously thought (here 1.96(8) Debye), and the T1 spin lifetime in low magnetic fields is longer than previously thought (> 4.6(1.5) hours). We furthermore determined the optical excited state lifetime (7.7(4) ns), and therefore the natural radiative efficiency (0.80(9) %), and by measuring the phonon sideband, determined the zero-phonon emission fraction (16(1) %). Taken together, these parameters indicate that an integrated quantum optoelectronic platform based upon chalcogen donor qubits in silicon is well within reach of current capabilities. △ Less

Submitted 26 September, 2018; originally announced September 2018.

Journal ref: Phys. Rev. Applied 11, 044036 (2019)

Highly enriched $^{28}$Si reveals remarkable optical linewidths and fine structure for well-known damage centers

Authors: C. Chartrand, L. Bergeron, K. J. Morse, H. Riemann, N. V. Abrosimov, P. Becker, H. -J. Pohl, S. Simmons, M. L. W. Thewalt

Abstract: Luminescence and optical absorption due to radiation damage centers in silicon has been studied exhaustively for decades, but is receiving new interest for applications as emitters for integrated silicon photonic technologies. While a variety of other optical transitions have been found to be much sharper in enriched $^{28}$Si than in natural Si, due to the elimination of inhomogeneous isotopic br… ▽ More Luminescence and optical absorption due to radiation damage centers in silicon has been studied exhaustively for decades, but is receiving new interest for applications as emitters for integrated silicon photonic technologies. While a variety of other optical transitions have been found to be much sharper in enriched $^{28}$Si than in natural Si, due to the elimination of inhomogeneous isotopic broadening, this has not yet been investigated for radiation damage centers. We report results for the well-known G, W and C damage centers in highly enriched $^{28}$Si, with optical linewidth improvements in some cases of over two orders of magnitude, revealing previously hidden fine structure in the G center emission and absorption. These results have direct implications for the linewidths to be expected from single center emission, even in natural Si, and for models for the G center structure. The advantages of $^{28}$Si can be readily extended to the study of other radiation damage centers in Si. △ Less

Submitted 27 July, 2018; originally announced July 2018.

Comments: 8 pages, 6 figures

Journal ref: Phys. Rev. B 98, 195201 (2018)

Further investigations of the deep double donor magnesium in silicon

Authors: Rohan Abraham, Adam DeAbreu, Kevin Morse, Valentina Shuman, Leonid Portsel, Anatoly Lodygin, Yuri Astrov, Nikolay Abrosimov, Sergey Pavlov, Heinz-Wilhelm Hübers, Stephanie Simmons, Michael Thewalt

Abstract: The deep double donor levels of substitutional chalcogen impurities in silicon have unique optical properties which may enable a spin/photonic quantum technology. The interstitial magnesium impurity (Mg$_i$) in silicon is also a deep double donor but has not yet been studied in the same detail as have the chalcogens. In this study we look at the neutral and singly ionized Mg$_i$ absorption spectra… ▽ More The deep double donor levels of substitutional chalcogen impurities in silicon have unique optical properties which may enable a spin/photonic quantum technology. The interstitial magnesium impurity (Mg$_i$) in silicon is also a deep double donor but has not yet been studied in the same detail as have the chalcogens. In this study we look at the neutral and singly ionized Mg$_i$ absorption spectra in natural silicon and isotopically enriched 28-silicon in more detail. The 1s(A$_1$) to 1s(T$_2$) transitions, which are very strong for the chalcogens and are central to the proposed spin/photonic quantum technology, could not be detected. We observe the presence of another double donor (Mg$_{i*}$) that may result from Mg$_i$ in a reduced symmetry configuration, most likely due to complexing with another impurity. The neutral species of Mg$_{i*}$ reveal unusual low lying ground state levels detected through temperature dependence studies. We also observe a shallow donor which we identify as a magnesium-boron pair. △ Less

Submitted 10 September, 2018; v1 submitted 5 June, 2018; originally announced June 2018.

Journal ref: Phys. Rev. B 98, 045202 (2018)

A simple random matrix model for the vibrational spectrum of jammed packings

Authors: E M Stanifer, P K Morse, A A Middleton, M L Manning

Abstract: To better understand the surprising low-frequency vibrational modes in structural glasses, we study the spectra of a large ensemble of sparse random matrices where disorder is controlled by the distribution of bond weights and network coordination. We find $D(ω)$ has three regimes: a very-low-frequency regime that can be predicted analytically using extremal statistics, an intermediate regime with… ▽ More To better understand the surprising low-frequency vibrational modes in structural glasses, we study the spectra of a large ensemble of sparse random matrices where disorder is controlled by the distribution of bond weights and network coordination. We find $D(ω)$ has three regimes: a very-low-frequency regime that can be predicted analytically using extremal statistics, an intermediate regime with quasi-localized modes, and a plateau with $D(ω) \sim ω^0$. In the special case of uniform bond weights, the intermediate regime displays $D(ω) \sim ω^4$, independent of network coordination and system size, just as recently discovered in simulations of structural glasses. △ Less

Submitted 27 April, 2018; v1 submitted 11 April, 2018; originally announced April 2018.

Journal ref: Phys. Rev. E 98, 042908 (2018)

Zero field optical magnetic resonance study of phosphorus donors in 28-silicon

Authors: Kevin J. Morse, Phillip Dluhy, Julian Huber, Jeff Z. Salvail, Kamyar Saeedi, Helge Riemann, Nikolay V. Abrosimov, Peter Becker, Hans-Joachim Pohl, S. Simmons, M. L. W. Thewalt

Abstract: Donor spins in silicon are some of the most promising qubits for upcoming solid-state quantum technologies. The nuclear spins of phosphorus donors in enriched silicon have among the longest coherence times of any solid-state system as well as simultaneous qubit initialization, manipulation and readout fidelities near ~99.9%. Here we characterize the phosphorus in silicon system in the regime of "z… ▽ More Donor spins in silicon are some of the most promising qubits for upcoming solid-state quantum technologies. The nuclear spins of phosphorus donors in enriched silicon have among the longest coherence times of any solid-state system as well as simultaneous qubit initialization, manipulation and readout fidelities near ~99.9%. Here we characterize the phosphorus in silicon system in the regime of "zero" magnetic field, where a singlet-triplet spin clock transition can be accessed, using laser spectroscopy and magnetic resonance methods. We show the system can be optically hyperpolarized and has ~10 s Hahn echo coherence times, even at Earth's magnetic field and below. △ Less

Submitted 30 January, 2018; originally announced January 2018.

Comments: 6 pages, 4 figures

Journal ref: Phys. Rev. B 97, 115205 (2018)

A photonic platform for donor spin qubits in silicon

Authors: Kevin J. Morse, Rohan J. S. Abraham, Helge Riemann, Nikolai V. Abrosimov, Peter Becker, Hans-Joachim Pohl, Michael L. W. Thewalt, Stephanie Simmons

Abstract: Donor impurity spins in silicon-28 are highly competitive qubits for upcoming solid-state quantum technologies, yet a proven scalable strategy for multi-qubit devices remains conspicuously absent. These CMOS-compatible, atomically identical qubits offer significant advantages including 3-hour coherence ($T_2$) lifetimes, as well as simultaneous qubit initialization, manipulation and readout fideli… ▽ More Donor impurity spins in silicon-28 are highly competitive qubits for upcoming solid-state quantum technologies, yet a proven scalable strategy for multi-qubit devices remains conspicuously absent. These CMOS-compatible, atomically identical qubits offer significant advantages including 3-hour coherence ($T_2$) lifetimes, as well as simultaneous qubit initialization, manipulation and readout fidelities near $\sim\!99.9\%$. These properties meet the requirements for many modern quantum error correction protocols, which are essential for constructing large-scale universal quantum technologies. However, a method of reliably coupling spatially-separated qubits, which crucially does not sacrifice qubit quality and is robust to manufacturing imperfections, has yet to be identified. Here we present such a platform for donor qubits in silicon, by exploiting optically-accessible `deep' chalcogen donors. We show that these donors emit highly uniform light, can be optically initialized, and offer long-lived spin qubit ground states without requiring milliKelvin temperatures. These combined properties make chalcogen donors uniquely suitable for incorporation into silicon photonic architectures for single-shot single-qubit readout as well as for multi-qubit coupling. This unlocks clear pathways for silicon-based quantum computing, spin to photon conversion, photonic memories, silicon-integrated triggered single photon sources and all-optical silicon switches. △ Less

Submitted 10 June, 2016; originally announced June 2016.

Comments: 7 pages, 4 figures

Hidden symmetries in jammed systems

Authors: Peter K. Morse, Eric I. Corwin

Abstract: There are deep, but hidden, geometric structures within jammed systems, associated with hidden symmetries. These can be revealed by repeated transformations under which these structures lead to fixed points. These geometric structures can be found in the Voronoi tesselation of space defined by the packing. In this paper we examine two iterative processes: maximum inscribed sphere (MIS) inversion a… ▽ More There are deep, but hidden, geometric structures within jammed systems, associated with hidden symmetries. These can be revealed by repeated transformations under which these structures lead to fixed points. These geometric structures can be found in the Voronoi tesselation of space defined by the packing. In this paper we examine two iterative processes: maximum inscribed sphere (MIS) inversion and a real-space coarsening scheme. Under repeated iterations of the MIS inversion process we find invariant systems in which every particle is equal to the maximum inscribed sphere within its Voronoi cell. Using a real-space coarsening scheme we reveal behavior in geometric order parameters which is length-scale invariant. △ Less

Submitted 14 April, 2016; v1 submitted 28 February, 2016; originally announced February 2016.

Geometric order parameters derived from the Voronoi tessellation show signatures of the jamming transition

Authors: Peter K. Morse, Eric I. Corwin

Abstract: A jammed packing of frictionless spheres at zero temperature is perfectly specified by the network of contact forces from which mechanical properties can be derived. However, we can alternatively consider a packing as a geometric structure, characterized by a Voronoi tessellation which encodes the local environment around each particle. We find that this local environment characterizes systems bot… ▽ More A jammed packing of frictionless spheres at zero temperature is perfectly specified by the network of contact forces from which mechanical properties can be derived. However, we can alternatively consider a packing as a geometric structure, characterized by a Voronoi tessellation which encodes the local environment around each particle. We find that this local environment characterizes systems both above and below jamming and changes markedly at the transition. A variety of order parameters derived from this tessellation carry signatures of the jamming transition, complete with scaling exponents. Furthermore, we define a real space geometric correlation function which also displays a signature of jamming. Taken together, these results demonstrate the validity and usefulness of a purely geometric approach to jamming. △ Less

Submitted 24 August, 2015; v1 submitted 28 April, 2015; originally announced April 2015.

Nodal quasiparticle dynamics in the heavy fermion superconductor CeCoIn5 revealed by precision microwave spectroscopy

Authors: C. J. S. Truncik, W. A. Huttema, P. J. Turner, S. Özcan, N. C. Murphy, P. R. Carrière, E. Thewalt, K. J. Morse, A. J. Koenig, J. L. Sarrao, D. M. Broun

Abstract: CeCoIn5 is a heavy fermion superconductor with strong similarities to the high-Tc cuprates, including quasi-two-dimensionality, proximity to antiferromagnetism, and probable d-wave pairing arising from a non-Fermi-liquid normal state. Experiments allowing detailed comparisons of their electronic properties are of particular interest, but in most cases are difficult to realize, due to their very di… ▽ More CeCoIn5 is a heavy fermion superconductor with strong similarities to the high-Tc cuprates, including quasi-two-dimensionality, proximity to antiferromagnetism, and probable d-wave pairing arising from a non-Fermi-liquid normal state. Experiments allowing detailed comparisons of their electronic properties are of particular interest, but in most cases are difficult to realize, due to their very different transition temperatures. Here we use low temperature microwave spectroscopy to study the charge dynamics of the CeCoIn5 superconducting state. The similarities to cuprates, in particular to ultra-clean YBa2Cu3Oy, are striking: the frequency and temperature dependence of the quasiparticle conductivity are instantly recognizable, a consequence of rapid suppression of quasiparticle scattering below Tc; and penetration depth data, when properly treated, reveal a clean, linear temperature dependence of the quasiparticle contribution to superfluid density. The measurements also expose key differences, including prominent multiband effects and a temperature dependent renormalization of the quasiparticle mass. △ Less

Submitted 22 September, 2013; v1 submitted 19 October, 2012; originally announced October 2012.

Comments: 9 pages, 6 figures

Journal ref: Nature Communications 4, 2477 (2013)

Probing the Sensitivity of Electron Wave Interference to Disorder-Induced Scattering in Solid-State Devices

Authors: B. C. Scannell, I. Pilgrim, A. M. See, R. D. Montgomery, P. K. Morse, M. S. Fairbanks, C. A. Marlow, H. Linke, I. Farrer, D. A. Ritchie, A. R. Hamilton, A. P. Micolich, L. Eaves, R. P. Taylor

Abstract: The study of electron motion in semiconductor billiards has elucidated our understanding of quantum interference and quantum chaos. The central assumption is that ionized donors generate only minor perturbations to the electron trajectories, which are determined by scattering from billiard walls. We use magnetoconductance fluctuations as a probe of the quantum interference and show that these fluc… ▽ More The study of electron motion in semiconductor billiards has elucidated our understanding of quantum interference and quantum chaos. The central assumption is that ionized donors generate only minor perturbations to the electron trajectories, which are determined by scattering from billiard walls. We use magnetoconductance fluctuations as a probe of the quantum interference and show that these fluctuations change radically when the scattering landscape is modified by thermally-induced charge displacement between donor sites. Our results challenge the accepted understanding of quantum interference effects in nanostructures. △ Less

Submitted 7 May, 2012; v1 submitted 28 June, 2011; originally announced June 2011.

Comments: 8 pages, 5 figures, Submitted to Physical Review B