Showing 1–31 of 31 results for author: Amin, K

Dynamics of Gate-Controlled Superconducting Dayem Bridges

Authors: François Joint, Kazi Rafsanjani Amin, Ivo Cools, Simone Gasparinetti

Abstract: Local control of superconducting circuits by high-impedance electrical gates offers potential advantages in superconducting logic, quantum processing units, and cryoelectronics. Recent experiments have reported gate-controlled supercurrent in Dayem bridges made of metallic superconductors, mediated by direct current leakage, out-of-equilibrium phonons, or possibly other mechanisms. However, a time… ▽ More Local control of superconducting circuits by high-impedance electrical gates offers potential advantages in superconducting logic, quantum processing units, and cryoelectronics. Recent experiments have reported gate-controlled supercurrent in Dayem bridges made of metallic superconductors, mediated by direct current leakage, out-of-equilibrium phonons, or possibly other mechanisms. However, a time-domain characterization of this effect has been lacking. Here, we integrate Dayem bridges made of Niobium on Silicon into coplanar-waveguide resonators, and measure the effect of the gate voltage at steady state and during pulsed operation. We consider two types of arrangements for the gate: a side-coupled gate and a remote injector. In both cases, we observe sizable changes in the real and the imaginary part of the constriction's impedance for gate voltages of the order of 1 V. However, we find striking differences in the time-domain dynamics, with the remote injector providing a faster and more controlled response. Our results contribute to our understanding of gate-controlled superconducting devices and their suitability for applications. △ Less

Submitted 12 May, 2024; originally announced May 2024.

Comments: 7 pages, 4 figures

First-order quantum breakdown of superconductivity in amorphous superconductors

Authors: Thibault Charpentier, David Perconte, Sébastien Léger, Kazi Rafsanjani Amin, Florent Blondelle, Frédéric Gay, Olivier Buisson, Lev Ioffe, Anton Khvalyuk, Igor Poboiko, Mikhail Feigel'man, Nicolas Roch, Benjamin Sacépé

Abstract: Continuous quantum phase transitions are widely assumed and frequently observed in various systems of quantum particles or spins. Their characteristic trait involves scaling laws governing a second-order, gradual suppression of the order parameter as the quantum critical point is approached. The localization of Cooper pairs in disordered superconductors and the resulting breakdown of superconducti… ▽ More Continuous quantum phase transitions are widely assumed and frequently observed in various systems of quantum particles or spins. Their characteristic trait involves scaling laws governing a second-order, gradual suppression of the order parameter as the quantum critical point is approached. The localization of Cooper pairs in disordered superconductors and the resulting breakdown of superconductivity have long stood as a prototypical example. Here, we show a departure from this paradigm, showcasing that amorphous superconducting films of indium oxide undergo a distinctive, discontinuous first-order quantum phase transition tuned by disorder. Through systematic measurements of the plasmon spectrum in superconducting microwave resonators, we provide evidence for a marked jump of both the zero-temperature superfluid stiffness and the transition temperature at the critical disorder. This discontinuous transition sheds light on the previously overlooked role of repulsive interactions between Cooper pairs and the subsequent competition between superconductivity and insulating Cooper-pair glass. Furthermore, our investigation shows that the critical temperature of the films no longer relates to the pairing amplitude but aligns with the superfluid stiffness, consistent with the pseudogap regime of preformed Cooper pairs. Our findings raise fundamental new questions into the role of disorder in quantum phase transitions and carry implications for superinductances in quantum circuits. △ Less

Submitted 27 May, 2024; v1 submitted 15 April, 2024; originally announced April 2024.

Comments: Updated references

Dynamical excitation control and multimode emission of an atom-photon bound state

Authors: Claudia Castillo-Moreno, Kazi Rafsanjani Amin, Ingrid Strandberg, Mikael Kervinen, Amr Osman, Simone Gasparinetti

Abstract: Atom-photon bound states arise from the coupling of quantum emitters to the band edge of dispersion-engineered waveguides. Thanks to their tunable-range interactions, they are promising building blocks for quantum simulators. Here, we study the dynamics of an atom-photon bound state emerging from coupling a frequency-tunable quantum emitter - a transmon-type superconducting circuit - to the band e… ▽ More Atom-photon bound states arise from the coupling of quantum emitters to the band edge of dispersion-engineered waveguides. Thanks to their tunable-range interactions, they are promising building blocks for quantum simulators. Here, we study the dynamics of an atom-photon bound state emerging from coupling a frequency-tunable quantum emitter - a transmon-type superconducting circuit - to the band edge of a microwave metamaterial. Employing precise temporal control over the frequency detuning of the emitter from the band edge, we examine the transition from adiabatic to non-adiabatic behavior in the formation of the bound state and its melting into the propagating modes of the metamaterial. Moreover, we experimentally observe multi-mode emission from the bound state, triggered by a fast change of the emitter's frequency. Our study offers insight into the dynamic preparation of atom-photon bound states and provides a method to characterize their photonic content, with implications in quantum optics and quantum simulation. △ Less

Submitted 8 April, 2024; originally announced April 2024.

Comments: 8 pages (4 figures) + supplementary information (5 pages)

Direct detection of quasiparticle tunneling with a charge-sensitive superconducting sensor coupled to a waveguide

Authors: Kazi Rafsanjani Amin, Axel M. Eriksson, Mikael Kervinen, Linus Andersson, Robert Rehammar, Simone Gasparinetti

Abstract: Detecting quasiparticle tunneling events in superconducting circuits provides information about the population and dynamics of non-equilibrium quasiparticles. Such events can be detected by monitoring changes in the frequency of an offset-charge-sensitive superconducting qubit. This monitoring has so far been performed by Ramsey interferometry assisted by a readout resonator. Here, we demonstrate… ▽ More Detecting quasiparticle tunneling events in superconducting circuits provides information about the population and dynamics of non-equilibrium quasiparticles. Such events can be detected by monitoring changes in the frequency of an offset-charge-sensitive superconducting qubit. This monitoring has so far been performed by Ramsey interferometry assisted by a readout resonator. Here, we demonstrate a quasiparticle detector based on a superconducting qubit directly coupled to a waveguide. We directly measure quasiparticle number parity on the qubit island by probing the coherent scattering of a microwave tone, offering simplicity of operation, fast detection speed, and a large signal-to-noise ratio. We observe tunneling rates between 0.8 and $7~\rm{s}^{-1}$, depending on the average occupation of the detector qubit, and achieve a temporal resolution below $10~μ\rm{s}$ without a quantum-limited amplifier. Our simple and efficient detector lowers the barrier to perform studies of quasiparticle population and dynamics, facilitating progress in fundamental science, quantum information processing, and sensing. △ Less

Submitted 1 April, 2024; originally announced April 2024.

Bengali License Plate Recognition: Unveiling Clarity with CNN and GFP-GAN

Authors: Noushin Afrin, Md Mahamudul Hasan, Mohammed Fazlay Elahi Safin, Khondakar Rifat Amin, Md Zahidul Haque, Farzad Ahmed, Md. Tanvir Rouf Shawon

Abstract: Automated License Plate Recognition(ALPR) is a system that automatically reads and extracts data from vehicle license plates using image processing and computer vision techniques. The Goal of LPR is to identify and read the license plate number accurately and quickly, even under challenging, conditions such as poor lighting, angled or obscured plates, and different plate fonts and layouts. The pro… ▽ More Automated License Plate Recognition(ALPR) is a system that automatically reads and extracts data from vehicle license plates using image processing and computer vision techniques. The Goal of LPR is to identify and read the license plate number accurately and quickly, even under challenging, conditions such as poor lighting, angled or obscured plates, and different plate fonts and layouts. The proposed method consists of processing the Bengali low-resolution blurred license plates and identifying the plate's characters. The processes include image restoration using GFPGAN, Maximizing contrast, Morphological image processing like dilation, feature extraction and Using Convolutional Neural Networks (CNN), character segmentation and recognition are accomplished. A dataset of 1292 images of Bengali digits and characters was prepared for this project. △ Less

Submitted 17 December, 2023; originally announced December 2023.

Universality of Anderson Localization Transitions in the Integer and Fractional Quantum Hall Regime

Authors: Simrandeep Kaur, Tanima Chanda, Kazi Rafsanjani Amin, Kenji Watanabe, Takashi Taniguchi, Unmesh Ghorai, Yuval Gefen, G. J. Sreejith, Aveek Bid

Abstract: Understanding the interplay between electronic interactions and disorder-induced localization has been a longstanding quest in the physics of quantum materials. One of the most convincing demonstrations of the scaling theory of localization for noninteracting electrons has come from plateau transitions in the integer quantum Hall effect with short-range disorder, wherein the localization length di… ▽ More Understanding the interplay between electronic interactions and disorder-induced localization has been a longstanding quest in the physics of quantum materials. One of the most convincing demonstrations of the scaling theory of localization for noninteracting electrons has come from plateau transitions in the integer quantum Hall effect with short-range disorder, wherein the localization length diverges as the critical filling factor is approached with a measured scaling exponent close to the theoretical estimates. In this work, we extend this physics to the fractional quantum Hall effect, a paradigmatic phenomenon arising from a confluence of interaction, disorder, and topology. We employ high-mobility trilayer graphene devices where the transport is dominated by short-range impurity scattering, and the extent of Landau level mixing can be varied by a perpendicular electric field. Our principal finding is that the plateau-to-plateau transitions from N+1/3 to N+2/5 and from N+2/5 to N+3/7 fractional states are governed by a universal scaling exponent, which is identical to that for the integer plateau transitions and is independent of the perpendicular electric field. These observations and the values of the critical filling factors are consistent with a description in terms of Anderson localization-delocalization transitions of weakly interacting electron-flux bound states called composite Fermions. This points to a universal effective physics underlying fractional and integer plateau-to-plateau transitions independent of the quasiparticle statistics of the phases and unaffected by weak Landau level mixing. Besides clarifying the conditions for the realization of the scaling regime for composite fermions, the work opens the possibility of exploring a wide variety of plateau transitions realized in graphene, including the fractional anomalous Hall phases and non-abelian FQH states. △ Less

Submitted 12 December, 2023; v1 submitted 11 December, 2023; originally announced December 2023.

ChatGPT-3.5, ChatGPT-4, Google Bard, and Microsoft Bing to Improve Health Literacy and Communication in Pediatric Populations and Beyond

Authors: Kanhai S. Amin, Linda Mayes, Pavan Khosla, Rushabh Doshi

Abstract: Purpose: Enhanced health literacy has been linked to better health outcomes; however, few interventions have been studied. We investigate whether large language models (LLMs) can serve as a medium to improve health literacy in children and other populations. Methods: We ran 288 conditions using 26 different prompts through ChatGPT-3.5, Microsoft Bing, and Google Bard. Given constraints imposed b… ▽ More Purpose: Enhanced health literacy has been linked to better health outcomes; however, few interventions have been studied. We investigate whether large language models (LLMs) can serve as a medium to improve health literacy in children and other populations. Methods: We ran 288 conditions using 26 different prompts through ChatGPT-3.5, Microsoft Bing, and Google Bard. Given constraints imposed by rate limits, we tested a subset of 150 conditions through ChatGPT-4. The primary outcome measurements were the reading grade level (RGL) and word counts of output. Results: Across all models, output for basic prompts such as "Explain" and "What is (are)" were at, or exceeded, a 10th-grade RGL. When prompts were specified to explain conditions from the 1st to 12th RGL, we found that LLMs had varying abilities to tailor responses based on RGL. ChatGPT-3.5 provided responses that ranged from the 7th-grade to college freshmen RGL while ChatGPT-4 outputted responses from the 6th-grade to the college-senior RGL. Microsoft Bing provided responses from the 9th to 11th RGL while Google Bard provided responses from the 7th to 10th RGL. Discussion: ChatGPT-3.5 and ChatGPT-4 did better in achieving lower-grade level outputs. Meanwhile Bard and Bing tended to consistently produce an RGL that is at the high school level regardless of prompt. Additionally, Bard's hesitancy in providing certain outputs indicates a cautious approach towards health information. LLMs demonstrate promise in enhancing health communication, but future research should verify the accuracy and effectiveness of such tools in this context. Implications: LLMs face challenges in crafting outputs below a sixth-grade reading level. However, their capability to modify outputs above this threshold provides a potential mechanism to improve health literacy and communication in a pediatric population and beyond. △ Less

Submitted 16 November, 2023; originally announced November 2023.

Comments: 15 pages, 1 Table, 3 Figures, and 3 Supplemental Figures

System-specific parameter optimization for non-polarizable and polarizable force fields

Authors: Xiaojuan Hu, Kazi S. Amin, Markus Schneider, Carmay Lim, Dennis Salahub, Carsten Baldauf

Abstract: The accuracy of classical force fields (FFs) has been shown to be limited for the simulation of cation-protein systems despite their importance in understanding the processes of life. Improvements can result from optimizing the parameters of classical FFs or by extending the FF formulation by terms describing charge transfer and polarization effects. In this work, we introduce our implementation o… ▽ More The accuracy of classical force fields (FFs) has been shown to be limited for the simulation of cation-protein systems despite their importance in understanding the processes of life. Improvements can result from optimizing the parameters of classical FFs or by extending the FF formulation by terms describing charge transfer and polarization effects. In this work, we introduce our implementation of the CTPOL model in OpenMM, which extends the classical additive FF formula by adding charge transfer (CT) and polarization (POL). Furthermore, we present an open-source parameterization tool, called FFAFFURR that enables the (system specific) parameterization of OPLS-AA and CTPOL models. The performance of our workflow was evaluated by its ability to reproduce quantum chemistry energies and by molecular dynamics simulations of a Zinc finger protein. △ Less

Submitted 9 October, 2023; v1 submitted 22 March, 2023; originally announced March 2023.

Comments: 62 pages and 25 figures (including SI), manuscript to be submitted soon

Comparing neural network training performance between Elixir and Python

Authors: Lucas C. Tavano, Lucas K. Amin, Adolfo Gustavo Serra-Seca-Neto

Abstract: With a wide range of libraries focused on the machine learning market, such as TensorFlow, NumPy, Pandas, Keras, and others, Python has made a name for itself as one of the main programming languages. In February 2021, José Valim and Sean Moriarity published the first version of the Numerical Elixir (Nx) library, a library for tensor operations written in Elixir. Nx aims to allow the language be a… ▽ More With a wide range of libraries focused on the machine learning market, such as TensorFlow, NumPy, Pandas, Keras, and others, Python has made a name for itself as one of the main programming languages. In February 2021, José Valim and Sean Moriarity published the first version of the Numerical Elixir (Nx) library, a library for tensor operations written in Elixir. Nx aims to allow the language be a good choice for GPU-intensive operations. This work aims to compare the results of Python and Elixir on training convolutional neural networks (CNN) using MNIST and CIFAR-10 datasets, concluding that Python achieved overall better results, and that Elixir is already a viable alternative. △ Less

Submitted 25 October, 2022; originally announced October 2022.

Learning-Augmented Private Algorithms for Multiple Quantile Release

Authors: Mikhail Khodak, Kareem Amin, Travis Dick, Sergei Vassilvitskii

Abstract: When applying differential privacy to sensitive data, we can often improve performance using external information such as other sensitive data, public data, or human priors. We propose to use the learning-augmented algorithms (or algorithms with predictions) framework -- previously applied largely to improve time complexity or competitive ratios -- as a powerful way of designing and analyzing priv… ▽ More When applying differential privacy to sensitive data, we can often improve performance using external information such as other sensitive data, public data, or human priors. We propose to use the learning-augmented algorithms (or algorithms with predictions) framework -- previously applied largely to improve time complexity or competitive ratios -- as a powerful way of designing and analyzing privacy-preserving methods that can take advantage of such external information to improve utility. This idea is instantiated on the important task of multiple quantile release, for which we derive error guarantees that scale with a natural measure of prediction quality while (almost) recovering state-of-the-art prediction-independent guarantees. Our analysis enjoys several advantages, including minimal assumptions about the data, a natural way of adding robustness, and the provision of useful surrogate losses for two novel ``meta" algorithms that learn predictions from other (potentially sensitive) data. We conclude with experiments on challenging tasks demonstrating that learning predictions across one or more instances can lead to large error reductions while preserving privacy. △ Less

Submitted 8 May, 2023; v1 submitted 20 October, 2022; originally announced October 2022.

Comments: To appear in ICML 2023

Easy Differentially Private Linear Regression

Authors: Kareem Amin, Matthew Joseph, Mónica Ribero, Sergei Vassilvitskii

Abstract: Linear regression is a fundamental tool for statistical analysis. This has motivated the development of linear regression methods that also satisfy differential privacy and thus guarantee that the learned model reveals little about any one data point used to construct it. However, existing differentially private solutions assume that the end user can easily specify good data bounds and hyperparame… ▽ More Linear regression is a fundamental tool for statistical analysis. This has motivated the development of linear regression methods that also satisfy differential privacy and thus guarantee that the learned model reveals little about any one data point used to construct it. However, existing differentially private solutions assume that the end user can easily specify good data bounds and hyperparameters. Both present significant practical obstacles. In this paper, we study an algorithm which uses the exponential mechanism to select a model with high Tukey depth from a collection of non-private regression models. Given $n$ samples of $d$-dimensional data used to train $m$ models, we construct an efficient analogue using an approximate Tukey depth that runs in time $O(d^2n + dm\log(m))$. We find that this algorithm obtains strong empirical performance in the data-rich setting with no data bounds or hyperparameter selection required. △ Less

Submitted 16 March, 2023; v1 submitted 15 August, 2022; originally announced August 2022.

Comments: This version corresponds to the camera-ready at ICLR 2023

A gate-tunable graphene Josephson parametric amplifier

Authors: Guilliam Butseraen, Arpit Ranadive, Nicolas Aparicio, Kazi Rafsanjani Amin, Abhishek Juyal, Martina Esposito, Kenji Watanabe, Takashi Taniguchi, Nicolas Roch, François Lefloch, Julien Renard

Abstract: With a large portfolio of elemental quantum components, superconducting quantum circuits have contributed to dramatic advances in microwave quantum optics. Of these elements, quantum-limited parametric amplifiers have proven to be essential for low noise readout of quantum systems whose energy range is intrinsically low (tens of $μ$eV ). They are also used to generate non classical states of light… ▽ More With a large portfolio of elemental quantum components, superconducting quantum circuits have contributed to dramatic advances in microwave quantum optics. Of these elements, quantum-limited parametric amplifiers have proven to be essential for low noise readout of quantum systems whose energy range is intrinsically low (tens of $μ$eV ). They are also used to generate non classical states of light that can be a resource for quantum enhanced detection. Superconducting parametric amplifiers, like quantum bits, typically utilize a Josephson junction as a source of magnetically tunable and dissipation-free nonlinearity. In recent years, efforts have been made to introduce semiconductor weak links as electrically tunable nonlinear elements, with demonstrations of microwave resonators and quantum bits using semiconductor nanowires, a two dimensional electron gas, carbon nanotubes and graphene. However, given the challenge of balancing nonlinearity, dissipation, participation, and energy scale, parametric amplifiers have not yet been implemented with a semiconductor weak link. Here we demonstrate a parametric amplifier leveraging a graphene Josephson junction and show that its working frequency is widely tunable with a gate voltage. We report gain exceeding 20 dB and noise performance close to the standard quantum limit. Our results complete the toolset for electrically tunable superconducting quantum circuits and offer new opportunities for the development of quantum technologies such as quantum computing, quantum sensing and fundamental science. △ Less

Submitted 5 April, 2022; originally announced April 2022.

Journal ref: Nature Nanotechnology (2022)

Plume: Differential Privacy at Scale

Authors: Kareem Amin, Jennifer Gillenwater, Matthew Joseph, Alex Kulesza, Sergei Vassilvitskii

Abstract: Differential privacy has become the standard for private data analysis, and an extensive literature now offers differentially private solutions to a wide variety of problems. However, translating these solutions into practical systems often requires confronting details that the literature ignores or abstracts away: users may contribute multiple records, the domain of possible records may be unknow… ▽ More Differential privacy has become the standard for private data analysis, and an extensive literature now offers differentially private solutions to a wide variety of problems. However, translating these solutions into practical systems often requires confronting details that the literature ignores or abstracts away: users may contribute multiple records, the domain of possible records may be unknown, and the eventual system must scale to large volumes of data. Failure to carefully account for all three issues can severely impair a system's quality and usability. We present Plume, a system built to address these problems. We describe a number of sometimes subtle implementation issues and offer practical solutions that, together, make an industrial-scale system for differentially private data analysis possible. Plume is currently deployed at Google and is routinely used to process datasets with trillions of records. △ Less

Submitted 27 January, 2022; originally announced January 2022.

Multifractal conductance fluctuations in high-mobility graphene in the Integer Quantum Hall regime

Authors: Kazi Rafsanjani Amin, Ramya Nagarajan, Rahul Pandit, Aveek Bid

Abstract: We present the first experimental evidence for the multifractality of a transport property at a topological phase transition. In particular, we show that conductance fluctuations display multifractality at the integer-quantum-Hall $ν=1 \longleftrightarrow ν=2$ plateau-to-plateau transition in a high-mobility mesoscopic graphene device. We establish that to observe this multifractality, it is cruci… ▽ More We present the first experimental evidence for the multifractality of a transport property at a topological phase transition. In particular, we show that conductance fluctuations display multifractality at the integer-quantum-Hall $ν=1 \longleftrightarrow ν=2$ plateau-to-plateau transition in a high-mobility mesoscopic graphene device. We establish that to observe this multifractality, it is crucial to work with very high-mobility devices with a well-defined critical point. This multifractality gets rapidly suppressed as the chemical potential moves away from these critical points. Our combination of multifractal analysis with state-of-the-art transport measurements at a topological phase transition provides a novel method for probing such phase transitions in mesoscopic devices. △ Less

Submitted 28 December, 2021; originally announced December 2021.

Comments: 10 pages, 4 figures

A CMOS compatible platform for high impedance superconducting quantum circuits

Authors: Kazi Rafsanjani Amin, Carine Ladner, Guillaume Jourdan, Sebastien Hentz, Nicolas Roch, Julien Renard

Abstract: Aluminium based platforms have allowed to reach major milestones for superconducting quantum circuits. For the next generation of devices, materials that are able to maintain low microwave losses while providing new functionalities, such as large kinetic inductance or compatibility with CMOS platform are sought for. Here we report on a combined direct current (DC) and microwave investigation of ti… ▽ More Aluminium based platforms have allowed to reach major milestones for superconducting quantum circuits. For the next generation of devices, materials that are able to maintain low microwave losses while providing new functionalities, such as large kinetic inductance or compatibility with CMOS platform are sought for. Here we report on a combined direct current (DC) and microwave investigation of titanium nitride lms of dierent thicknesses grown using CMOS compatible methods. For microwave resonators made of TiN lm of thickness $\sim$3 nm, we measured large kinetic inductance LK $\sim$ 240 pH/sq, high mode impedance of $\sim$ 4.2 k$Ω$ while maintaining microwave quality factor $\sim$ 10^5 in the single photon limit. We present an in-depth study of the microwave loss mechanisms in these devices that indicates the importance of quasiparticles and provide insights for further improvement. △ Less

Submitted 14 September, 2021; originally announced September 2021.

Journal ref: Appl. Phys. Lett. 120, 164001 (2022)

Learning with User-Level Privacy

Authors: Daniel Levy, Ziteng Sun, Kareem Amin, Satyen Kale, Alex Kulesza, Mehryar Mohri, Ananda Theertha Suresh

Abstract: We propose and analyze algorithms to solve a range of learning tasks under user-level differential privacy constraints. Rather than guaranteeing only the privacy of individual samples, user-level DP protects a user's entire contribution ($m \ge 1$ samples), providing more stringent but more realistic protection against information leaks. We show that for high-dimensional mean estimation, empirical… ▽ More We propose and analyze algorithms to solve a range of learning tasks under user-level differential privacy constraints. Rather than guaranteeing only the privacy of individual samples, user-level DP protects a user's entire contribution ($m \ge 1$ samples), providing more stringent but more realistic protection against information leaks. We show that for high-dimensional mean estimation, empirical risk minimization with smooth losses, stochastic convex optimization, and learning hypothesis classes with finite metric entropy, the privacy cost decreases as $O(1/\sqrt{m})$ as users provide more samples. In contrast, when increasing the number of users $n$, the privacy cost decreases at a faster $O(1/n)$ rate. We complement these results with lower bounds showing the minimax optimality of our algorithms for mean estimation and stochastic convex optimization. Our algorithms rely on novel techniques for private mean estimation in arbitrary dimension with error scaling as the concentration radius $τ$ of the distribution rather than the entire range. △ Less

Submitted 3 December, 2021; v1 submitted 23 February, 2021; originally announced February 2021.

Comments: NeurIPS 2021. 43 pages, 0 figure

Effect of dimensionality on the vortex-dynamics in type-II superconductor

Authors: Hemanta Kumar Kundu, Kazi Rafsanjani Amin, John Jesudasan, Pratap Raychaudhuri, Subroto Mukerjee, Aveek Bid

Abstract: We explore the effects of sample dimensionality on vortex pinning in a type-II, low-$T_C$, s-wave superconductor, NbN, in the presence of a perpendicular magnetic field, $H$. We find significant differences in the phase diagrams in the magnetic field--temperature plane between 3-dimensional (3D) and 2-dimensional (2D) NbN films. The differences are most striking close to the normal-superconductor… ▽ More We explore the effects of sample dimensionality on vortex pinning in a type-II, low-$T_C$, s-wave superconductor, NbN, in the presence of a perpendicular magnetic field, $H$. We find significant differences in the phase diagrams in the magnetic field--temperature plane between 3-dimensional (3D) and 2-dimensional (2D) NbN films. The differences are most striking close to the normal-superconductor phase transition. We establish that these variances have their origin in the differing pinning properties in two different dimensions. We obtain the pinning strength quantitatively in both the dimensions from two independent transport measurements performed in two different regimes of vortex-motion -- (i) thermally assisted flux-flow (TAFF) regime and (ii) flux flow (FF) regime. Both the measurements consistently show that both the pinning potential and the zero-field free-energy barrier to depinning in the 3D superconductor are at least an order of magnitude stronger than that in the 2D superconductor. Further, we probed the dynamics of pinning in both 2D and 3D superconductor through voltage fluctuation spectroscopy. We find that the mechanism of vortex pinning-depinning is qualitatively similar for the 3D and 2D superconductors. The voltage-fluctuations arising from vortex-motion are found to be correlated only in the 2D superconductor. We establish this to be due to the presence of long-range phase fluctuations near the Berezinskii-Kosterlitz-Thouless (BKT) type superconducting transition in 2-dimensional superconductors. △ Less

Submitted 19 September, 2020; originally announced September 2020.

Pan-Private Uniformity Testing

Authors: Kareem Amin, Matthew Joseph, Jieming Mao

Abstract: A centrally differentially private algorithm maps raw data to differentially private outputs. In contrast, a locally differentially private algorithm may only access data through public interaction with data holders, and this interaction must be a differentially private function of the data. We study the intermediate model of pan-privacy. Unlike a locally private algorithm, a pan-private algorithm… ▽ More A centrally differentially private algorithm maps raw data to differentially private outputs. In contrast, a locally differentially private algorithm may only access data through public interaction with data holders, and this interaction must be a differentially private function of the data. We study the intermediate model of pan-privacy. Unlike a locally private algorithm, a pan-private algorithm receives data in the clear. Unlike a centrally private algorithm, the algorithm receives data one element at a time and must maintain a differentially private internal state while processing this stream. First, we show that pure pan-privacy against multiple intrusions on the internal state is equivalent to sequentially interactive local privacy. Next, we contextualize pan-privacy against a single intrusion by analyzing the sample complexity of uniformity testing over domain $[k]$. Focusing on the dependence on $k$, centrally private uniformity testing has sample complexity $Θ(\sqrt{k})$, while noninteractive locally private uniformity testing has sample complexity $Θ(k)$. We show that the sample complexity of pure pan-private uniformity testing is $Θ(k^{2/3})$. By a new $Ω(k)$ lower bound for the sequentially interactive setting, we also separate pan-private from sequentially interactive locally private and multi-intrusion pan-private uniformity testing. △ Less

Submitted 17 July, 2020; v1 submitted 4 November, 2019; originally announced November 2019.

Exotic Multifractal Conductance Fluctuations in Graphene

Authors: Kazi Rafsanjani Amin, Samriddhi Sankar Ray, Nairita Pal, Rahul Pandit, Aveek Bid

Abstract: In quantum systems, signatures of multifractality are rare. They have been found only in the multiscaling of eigenfunctions at critical points. Here we demonstrate multifractality in the magnetic-field-induced universal conductance fluctuations of the conductance in a quantum condensed-matter system, namely, high-mobility single-layer graphene field-effect transistors. This multifractality decreas… ▽ More In quantum systems, signatures of multifractality are rare. They have been found only in the multiscaling of eigenfunctions at critical points. Here we demonstrate multifractality in the magnetic-field-induced universal conductance fluctuations of the conductance in a quantum condensed-matter system, namely, high-mobility single-layer graphene field-effect transistors. This multifractality decreases as the temperature increases or as do** moves the system away from the Dirac point. Our measurements and analysis present evidence for an incipient Anderson-localization near the Dirac point as the most plausible cause for this multifractality. Our experiments suggest that multifractality in the scaling behaviour of local eigenfunctions are reflected in macroscopic transport coefficients. We conjecture that an incipient Anderson-localization transition may be the origin of this multifractality. It is possible that multifractality is ubiquitous in transport properties of low-dimensional systems. Indeed, our work suggests that we should look for multifractality in transport in other low-dimensional quantum condensed-matter systems. △ Less

Submitted 12 April, 2018; originally announced April 2018.

Journal ref: Communications Physics, volume 1, Article number: 1 (2018)

Towards a Crowd Analytic Framework For Crowd Management in Majid-al-Haram

Authors: Sultan Daud Khan, Muhammad Tayyab, Muhammad Khurram Amin, Akram Nour, Anas Basalamah, Saleh Basalamah, Sohaib Ahmad Khan

Abstract: The scared cities of Makkah Al Mukarramah and Madina Al Munawarah host millions of pilgrims every year. During Hajj, the movement of large number of people has a unique spatial and temporal constraints, which makes Hajj one of toughest challenges for crowd management. In this paper, we propose a computer vision based framework that automatically analyses video sequence and computes important measu… ▽ More The scared cities of Makkah Al Mukarramah and Madina Al Munawarah host millions of pilgrims every year. During Hajj, the movement of large number of people has a unique spatial and temporal constraints, which makes Hajj one of toughest challenges for crowd management. In this paper, we propose a computer vision based framework that automatically analyses video sequence and computes important measurements which include estimation of crowd density, identification of dominant patterns, detection and localization of congestion. In addition, we analyze helpful statistics of the crowd like speed, and direction, that could provide support to crowd management personnel. The framework presented in this paper indicate that new advances in computer vision and machine learning can be leveraged effectively for challenging and high density crowd management applications. However, significant customization of existing approaches is required to apply them to the challenging crowd management situations in Masjid Al Haram. Our results paint a promising picture for deployment of computer vision technologies to assist in quantitative measurement of crowd size, density and congestion. △ Less

Submitted 14 September, 2017; originally announced September 2017.

Comments: 17th Scientific Meeting on Hajj & Umrah Research, 2017

Repeated Inverse Reinforcement Learning

Authors: Kareem Amin, Nan Jiang, Satinder Singh

Abstract: We introduce a novel repeated Inverse Reinforcement Learning problem: the agent has to act on behalf of a human in a sequence of tasks and wishes to minimize the number of tasks that it surprises the human by acting suboptimally with respect to how the human would have acted. Each time the human is surprised, the agent is provided a demonstration of the desired behavior by the human. We formalize… ▽ More We introduce a novel repeated Inverse Reinforcement Learning problem: the agent has to act on behalf of a human in a sequence of tasks and wishes to minimize the number of tasks that it surprises the human by acting suboptimally with respect to how the human would have acted. Each time the human is surprised, the agent is provided a demonstration of the desired behavior by the human. We formalize this problem, including how the sequence of tasks is chosen, in a few different ways and provide some foundational results. △ Less

Submitted 3 November, 2017; v1 submitted 15 May, 2017; originally announced May 2017.

Comments: The first two authors contributed equally to this work. The paper appears in NIPS 2017

Towards Resolving Unidentifiability in Inverse Reinforcement Learning

Authors: Kareem Amin, Satinder Singh

Abstract: We consider a setting for Inverse Reinforcement Learning (IRL) where the learner is extended with the ability to actively select multiple environments, observing an agent's behavior on each environment. We first demonstrate that if the learner can experiment with any transition dynamics on some fixed set of states and actions, then there exists an algorithm that reconstructs the agent's reward fun… ▽ More We consider a setting for Inverse Reinforcement Learning (IRL) where the learner is extended with the ability to actively select multiple environments, observing an agent's behavior on each environment. We first demonstrate that if the learner can experiment with any transition dynamics on some fixed set of states and actions, then there exists an algorithm that reconstructs the agent's reward function to the fullest extent theoretically possible, and that requires only a small (logarithmic) number of experiments. We contrast this result to what is known about IRL in single fixed environments, namely that the true reward function is fundamentally unidentifiable. We then extend this setting to the more realistic case where the learner may not select any transition dynamic, but rather is restricted to some fixed set of environments that it may try. We connect the problem of maximizing the information derived from experiments to submodular function maximization and demonstrate that a greedy algorithm is near optimal (up to logarithmic factors). Finally, we empirically validate our algorithm on an environment inspired by behavioral psychology. △ Less

Submitted 25 January, 2016; originally announced January 2016.

High performance sensors based on resistance fluctuations of single layer graphene transistors

Authors: Kazi Rafsanjani Amin, Aveek Bid

Abstract: One of the most interesting predicted applications of graphene monolayer based devices is as high quality sensors. In this letter we show, through systematic experiments, a chemical vapor sensor based on the measurement of low frequency resistance fluctuations of single layer graphene field-effect-transistor (SLG-FET) devices. The sensor has extremely high sensitivity, very high specificity, high… ▽ More One of the most interesting predicted applications of graphene monolayer based devices is as high quality sensors. In this letter we show, through systematic experiments, a chemical vapor sensor based on the measurement of low frequency resistance fluctuations of single layer graphene field-effect-transistor (SLG-FET) devices. The sensor has extremely high sensitivity, very high specificity, high fidelity and fast response times. The performance of the device using this scheme of measurement (which uses resistance fluctuations as the detection parameter) is more than two orders of magnitude better than a detection scheme where changes in the average value of the resistance is monitored. We propose a number-density fluctuation based model to explain the superior characteristics of noise measurement based detection scheme presented in this article. △ Less

Submitted 19 November, 2015; originally announced November 2015.

Journal ref: ACS Applied Materials and Interfaces, 2015, 7(35)

Role of different scattering mechanisms on the temperature dependence of transport in graphene

Authors: Suman Sarkar, Kazi Rafsanjani Amin, Ranjan Modak, Amandeep Singh, Subroto Mukerjee, Aveek Bid

Abstract: Detailed experimental and theoretical studies of the temperature dependence of the effect of different scattering mechanisms on electrical transport properties of graphene devices are presented. We find that for high mobility devices the transport properties are mainly governed by completely screened short range impurity scattering. On the other hand, for the low mobility devices transport propert… ▽ More Detailed experimental and theoretical studies of the temperature dependence of the effect of different scattering mechanisms on electrical transport properties of graphene devices are presented. We find that for high mobility devices the transport properties are mainly governed by completely screened short range impurity scattering. On the other hand, for the low mobility devices transport properties are determined by both types of scattering potentials - long range due to ionized impurities and short range due to completely screened charged impurities. The results could be explained in the framework of Boltzmann transport equations involving the two independent scattering mechanisms. △ Less

Submitted 28 October, 2015; originally announced October 2015.

Comments: Accepted for Publication in Scientific Reports, 2015

Journal ref: Scientific Reports 5, Article number: 16772 (2015)

Effect of ambient on the resistance fluctuations of graphene

Authors: Kazi Rafsanjani Amin, Aveek Bid

Abstract: In this letter we present the results of systematic experimental investigations of the effect of different chemical environments on the low frequency resistance fluctuations of single layer graphene field effect transistors (SLG-FET). The shape of the power spectral density of noise was found to be determined by the energetics of the adsorption-desorption of molecules from the graphene surface mak… ▽ More In this letter we present the results of systematic experimental investigations of the effect of different chemical environments on the low frequency resistance fluctuations of single layer graphene field effect transistors (SLG-FET). The shape of the power spectral density of noise was found to be determined by the energetics of the adsorption-desorption of molecules from the graphene surface making it the dominant source of noise in these devices. We also demonstrate a method of quantitatively determining the adsorption energies of chemicals on graphene surface based on noise measurements. We find that the magnitude of noise is extremely sensitive to the nature and amount of the chemical species present. We propose that a chemical sensor based on the measurement of low frequency resistance fluctuations of single layer graphene field effect transistor devices will have extremely high sensitivity, very high specificity, high fidelity and fast response times. △ Less

Submitted 28 October, 2015; originally announced October 2015.

Journal ref: Applied Physics Letters. 106, 183105 (2015)

Robust local and non-local transport in the Topological Kondo Insulator SmB$_{6}$ in the presence of high magnetic field

Authors: Sangram Biswas, Ramya Nagarajan, Suman Sarkar, Kazi Rafsanjani Amin, M. Ciomaga Hatnean, S. Tewari, G. Balakrishnan, Aveek Bid

Abstract: SmB$_6$ has been predicted to be a Kondo Topological Insulator with topologically protected conducting surface states. We have studied quantitatively the electrical transport through surface states in high quality single crystals of SmB$_6$. We observe a large non-local surface signal at temperatures lower than the bulk Kondo gap scale. Measurements and finite element simulations allow us to disti… ▽ More SmB$_6$ has been predicted to be a Kondo Topological Insulator with topologically protected conducting surface states. We have studied quantitatively the electrical transport through surface states in high quality single crystals of SmB$_6$. We observe a large non-local surface signal at temperatures lower than the bulk Kondo gap scale. Measurements and finite element simulations allow us to distinguish unambiguously between the contributions from different transport channels. In contrast to general expectations, the electrical transport properties of the surface channels was found to be insensitive to high magnetic fields. Local and non-local magnetoresistance measurements allowed us to identify definite signatures of helical spin states and strong inter-band scattering at the surface. △ Less

Submitted 11 February, 2015; originally announced February 2015.

Comments: 7 pages, 8 figures, 1 table

Online Learning and Profit Maximization from Revealed Preferences

Authors: Kareem Amin, Rachel Cummings, Lili Dworkin, Michael Kearns, Aaron Roth

Abstract: We consider the problem of learning from revealed preferences in an online setting. In our framework, each period a consumer buys an optimal bundle of goods from a merchant according to her (linear) utility function and current prices, subject to a budget constraint. The merchant observes only the purchased goods, and seeks to adapt prices to optimize his profits. We give an efficient algorithm fo… ▽ More We consider the problem of learning from revealed preferences in an online setting. In our framework, each period a consumer buys an optimal bundle of goods from a merchant according to her (linear) utility function and current prices, subject to a budget constraint. The merchant observes only the purchased goods, and seeks to adapt prices to optimize his profits. We give an efficient algorithm for the merchant's problem that consists of a learning phase in which the consumer's utility function is (perhaps partially) inferred, followed by a price optimization step. We also consider an alternative online learning algorithm for the setting where prices are set exogenously, but the merchant would still like to predict the bundle that will be bought by the consumer for purposes of inventory or supply chain management. In contrast with most prior work on the revealed preferences problem, we demonstrate that by making stronger assumptions on the form of utility functions, efficient algorithms for both learning and profit maximization are possible, even in adaptive, online settings. △ Less

Submitted 28 November, 2014; v1 submitted 27 July, 2014; originally announced July 2014.

Learning Prices for Repeated Auctions with Strategic Buyers

Authors: Kareem Amin, Afshin Rostamizadeh, Umar Syed

Abstract: Inspired by real-time ad exchanges for online display advertising, we consider the problem of inferring a buyer's value distribution for a good when the buyer is repeatedly interacting with a seller through a posted-price mechanism. We model the buyer as a strategic agent, whose goal is to maximize her long-term surplus, and we are interested in mechanisms that maximize the seller's long-term reve… ▽ More Inspired by real-time ad exchanges for online display advertising, we consider the problem of inferring a buyer's value distribution for a good when the buyer is repeatedly interacting with a seller through a posted-price mechanism. We model the buyer as a strategic agent, whose goal is to maximize her long-term surplus, and we are interested in mechanisms that maximize the seller's long-term revenue. We define the natural notion of strategic regret --- the lost revenue as measured against a truthful (non-strategic) buyer. We present seller algorithms that are no-(strategic)-regret when the buyer discounts her future surplus --- i.e. the buyer prefers showing advertisements to users sooner rather than later. We also give a lower bound on strategic regret that increases as the buyer's discounting weakens and shows, in particular, that any seller algorithm will suffer linear strategic regret if there is no discounting. △ Less

Submitted 26 November, 2013; originally announced November 2013.

Comments: Neural Information Processing Systems (NIPS 2013)

Probing long-range correlations in the Berezinskii-Kosterlitz-Thouless fluctuation regime of ultra-thin NbN superconducting films using transport noise measurements

Authors: R. Koushik, Siddhartha Kumar, Kazi Rafsanjani Amin, Mintu Mondal, John Jesudasan, Aveek Bid, Pratap Raychaudhuri, Arindam Ghosh

Abstract: We probe the presence of long-range correlations in phase fluctuations by analyzing the higher-order spectrum of resistance fluctuations in ultra-thin NbN superconducting films. The non-Gaussian component of resistance fluctuations is found to be sensitive to film thickness close to the transition, which allows us to distinguish between mean field and Berezinskii-Kosterlitz-Thouless (BKT) type sup… ▽ More We probe the presence of long-range correlations in phase fluctuations by analyzing the higher-order spectrum of resistance fluctuations in ultra-thin NbN superconducting films. The non-Gaussian component of resistance fluctuations is found to be sensitive to film thickness close to the transition, which allows us to distinguish between mean field and Berezinskii-Kosterlitz-Thouless (BKT) type superconducting transitions. The extent of non-Gaussianity was found to be bounded by the BKT and mean field transition temperatures and depend strongly on the roughness and structural inhomogeneity of the superconducting films. Our experiment outlines a novel fluctuation-based kinetic probe in detecting the nature of superconductivity in disordered low-dimensional materials. △ Less

Submitted 24 August, 2013; v1 submitted 20 August, 2013; originally announced August 2013.

Comments: submitted to PRL

Budget Optimization for Sponsored Search: Censored Learning in MDPs

Authors: Kareem Amin, Michael Kearns, Peter Key, Anton Schwaighofer

Abstract: We consider the budget optimization problem faced by an advertiser participating in repeated sponsored search auctions, seeking to maximize the number of clicks attained under that budget. We cast the budget optimization problem as a Markov Decision Process (MDP) with censored observations, and propose a learning algorithm based on the wellknown Kaplan-Meier or product-limit estimator. We validate… ▽ More We consider the budget optimization problem faced by an advertiser participating in repeated sponsored search auctions, seeking to maximize the number of clicks attained under that budget. We cast the budget optimization problem as a Markov Decision Process (MDP) with censored observations, and propose a learning algorithm based on the wellknown Kaplan-Meier or product-limit estimator. We validate the performance of this algorithm by comparing it to several others on a large set of search auction data from Microsoft adCenter, demonstrating fast convergence to optimal performance. △ Less

Submitted 16 October, 2012; originally announced October 2012.

Comments: Appears in Proceedings of the Twenty-Eighth Conference on Uncertainty in Artificial Intelligence (UAI2012)

Report number: UAI-P-2012-PG-54-63

Graphical Models for Bandit Problems

Authors: Kareem Amin, Michael Kearns, Umar Syed

Abstract: We introduce a rich class of graphical models for multi-armed bandit problems that permit both the state or context space and the action space to be very large, yet succinctly specify the payoffs for any context-action pair. Our main result is an algorithm for such models whose regret is bounded by the number of parameters and whose running time depends only on the treewidth of the graph substruct… ▽ More We introduce a rich class of graphical models for multi-armed bandit problems that permit both the state or context space and the action space to be very large, yet succinctly specify the payoffs for any context-action pair. Our main result is an algorithm for such models whose regret is bounded by the number of parameters and whose running time depends only on the treewidth of the graph substructure induced by the action space. △ Less

Submitted 14 February, 2012; originally announced February 2012.

Report number: UAI-P-2011-PG-1-10