Showing 1–10 of 10 results for author: Gee, A

Quantum Interference Enhances the Performance of Single-Molecule Transistors

Authors: Zhixin Chen, Iain M. Grace, Steffen L. Woltering, Lina Chen, Alex Gee, Jonathan Baugh, G. Andrew D. Briggs, Lapo Bogani, Jan A. Mol, Colin J. Lambert, Harry L. Anderson, James O. Thomas

Abstract: An unresolved challenge facing electronics at a few-nm scale is that resistive channels start leaking due to quantum tunneling. This affects the performance of nanoscale transistors, with single-molecule devices displaying particularly low switching ratios and operating frequencies, combined with large subthreshold swings.1 The usual strategy to mitigate quantum effects has been to increase device… ▽ More An unresolved challenge facing electronics at a few-nm scale is that resistive channels start leaking due to quantum tunneling. This affects the performance of nanoscale transistors, with single-molecule devices displaying particularly low switching ratios and operating frequencies, combined with large subthreshold swings.1 The usual strategy to mitigate quantum effects has been to increase device complexity, but theory shows that if quantum effects are exploited correctly, they can simultaneously lower energy consumption and boost device performance.2-6 Here, we demonstrate experimentally how the performance of molecular transistors can be improved when the resistive channel contains two destructively-interfering waves. We use a zinc-porphyrin coupled to graphene electrodes in a three-terminal transistor device to demonstrate a >104 conductance-switching ratio, a subthreshold swing at the thermionic limit, a > 7 kHz operating frequency, and stability over >105 cycles. This performance is competitive with the best nanoelectronic transistors. We fully map the antiresonance interference features in conductance, reproduce the behaviour by density functional theory calculations, and trace back this high performance to the coupling between molecular orbitals and graphene edge states. These results demonstrate how the quantum nature of electron transmission at the nanoscale can enhance, rather than degrade, device performance, and highlight directions for future development of miniaturised electronics. △ Less

Submitted 17 April, 2023; originally announced April 2023.

Comments: 11 pages, 4 figures

FASTER-CE: Fast, Sparse, Transparent, and Robust Counterfactual Explanations

Authors: Shubham Sharma, Alan H. Gee, Jette Henderson, Joydeep Ghosh

Abstract: Counterfactual explanations have substantially increased in popularity in the past few years as a useful human-centric way of understanding individual black-box model predictions. While several properties desired of high-quality counterfactuals have been identified in the literature, three crucial concerns: the speed of explanation generation, robustness/sensitivity and succinctness of explanation… ▽ More Counterfactual explanations have substantially increased in popularity in the past few years as a useful human-centric way of understanding individual black-box model predictions. While several properties desired of high-quality counterfactuals have been identified in the literature, three crucial concerns: the speed of explanation generation, robustness/sensitivity and succinctness of explanations (sparsity) have been relatively unexplored. In this paper, we present FASTER-CE: a novel set of algorithms to generate fast, sparse, and robust counterfactual explanations. The key idea is to efficiently find promising search directions for counterfactuals in a latent space that is specified via an autoencoder. These directions are determined based on gradients with respect to each of the original input features as well as of the target, as estimated in the latent space. The ability to quickly examine combinations of the most promising gradient directions as well as to incorporate additional user-defined constraints allows us to generate multiple counterfactual explanations that are sparse, realistic, and robust to input manipulations. Through experiments on three datasets of varied complexities, we show that FASTER-CE is not only much faster than other state of the art methods for generating multiple explanations but also is significantly superior when considering a larger set of desirable (and often conflicting) properties. Specifically we present results across multiple performance metrics: sparsity, proximity, validity, speed of generation, and the robustness of explanations, to highlight the capabilities of the FASTER-CE family. △ Less

Submitted 12 October, 2022; originally announced October 2022.

Localised Solid-State Nanopore Fabrication via Controlled Breakdown using On-Chip Electrodes

Authors: Jasper P. Fried, Jacob L. Swett, Binoy Paulose Nadappuram, Aleksandra Fedosyuk, Alex Gee, Ondrej E. Dyck, James R. Yates, Aleksandar P. Ivanov, Joshua B. Edel, Jan A. Mol

Abstract: Controlled breakdown has recently emerged as a highly accessible technique to fabricate solid-state nanopores. However, in its most common form, controlled breakdown creates a single nanopore at an arbitrary location in the membrane. Here, we introduce a new strategy whereby breakdown is performed by applying the electric field between an on-chip electrode and an electrolyte solution in contact wi… ▽ More Controlled breakdown has recently emerged as a highly accessible technique to fabricate solid-state nanopores. However, in its most common form, controlled breakdown creates a single nanopore at an arbitrary location in the membrane. Here, we introduce a new strategy whereby breakdown is performed by applying the electric field between an on-chip electrode and an electrolyte solution in contact with the opposite side of the membrane. We demonstrate two advantages of this method. First, we can independently fabricate multiple nanopores at given positions in the membrane by localising the applied field to the electrode. Second, we show we can create nanopores that are self-aligned with complementary nanoelectrodes by applying voltages to the on-chip electrodes to locally heat the membrane during controlled breakdown. This new controlled breakdown method provides a path towards the affordable, rapid, and automatable fabrication of arrays of nanopores self-aligned with complementary on-chip nanostructures. △ Less

Submitted 4 November, 2021; originally announced November 2021.

Statistical analysis of spin switching in coupled spin-crossover molecules

Authors: Philipp Stegmann, Alex Gee, Neil T. Kemp, Jürgen König

Abstract: We study the switching behavior of two spin-crossover molecules residing in a nanojunction device consisting of two closely spaced gold electrodes. The spin states are monitored through a real-time measurement of the resistance of the junction. A statistical analysis of the resistance values, the occupation probabilities, and the lifetimes of the respective spin states shows that the two spin-cros… ▽ More We study the switching behavior of two spin-crossover molecules residing in a nanojunction device consisting of two closely spaced gold electrodes. The spin states are monitored through a real-time measurement of the resistance of the junction. A statistical analysis of the resistance values, the occupation probabilities, and the lifetimes of the respective spin states shows that the two spin-crossover molecules are coupled to each other. We extract the parameters for a minimal model describing the two coupled spin-crossover molecules. Finally, we use the time dependence of factorial cumulants to demonstrate that the measured data indicates the presence of interactions between the two spin-crossover molecules. △ Less

Submitted 25 July, 2021; originally announced July 2021.

Comments: 9 pages, 5 figures

Journal ref: Phys. Rev. B 104, 125431 (2021)

FaiR-N: Fair and Robust Neural Networks for Structured Data

Authors: Shubham Sharma, Alan H. Gee, David Paydarfar, Joydeep Ghosh

Abstract: Fairness in machine learning is crucial when individuals are subject to automated decisions made by models in high-stake domains. Organizations that employ these models may also need to satisfy regulations that promote responsible and ethical A.I. While fairness metrics relying on comparing model error rates across subpopulations have been widely investigated for the detection and mitigation of bi… ▽ More Fairness in machine learning is crucial when individuals are subject to automated decisions made by models in high-stake domains. Organizations that employ these models may also need to satisfy regulations that promote responsible and ethical A.I. While fairness metrics relying on comparing model error rates across subpopulations have been widely investigated for the detection and mitigation of bias, fairness in terms of the equalized ability to achieve recourse for different protected attribute groups has been relatively unexplored. We present a novel formulation for training neural networks that considers the distance of data points to the decision boundary such that the new objective: (1) reduces the average distance to the decision boundary between two groups for individuals subject to a negative outcome in each group, i.e. the network is more fair with respect to the ability to obtain recourse, and (2) increases the average distance of data points to the boundary to promote adversarial robustness. We demonstrate that training with this loss yields more fair and robust neural networks with similar accuracies to models trained without it. Moreover, we qualitatively motivate and empirically show that reducing recourse disparity across groups also improves fairness measures that rely on error rates. To the best of our knowledge, this is the first time that recourse capabilities across groups are considered to train fairer neural networks, and a relation between error rates based fairness and recourse based fairness is investigated. △ Less

Submitted 12 October, 2020; originally announced October 2020.

Multilevel Resistance Switching and Enhanced Spin Transition Temperature in Single Molecule Spin Crossover Nanogap Devices

Authors: Alex Gee, Ayoub H. Jaafar, Barbora Brachňaková, Jamie Massey, Christopher H. Marrows, Ivan Šalitroš, N. T. Kemp

Abstract: Spin crossover (SCO) molecules are promising bi-stable magnetic switches with applications in molecular spintronics. However, little is known about the switching effects of a single SCO molecule when it is confined between two metal electrodes. Here we examine the switching properties of a [Fe(III)(EtOSalPet )(NCS)] SCO molecule that is specifically tailored for surface deposition and binding to o… ▽ More Spin crossover (SCO) molecules are promising bi-stable magnetic switches with applications in molecular spintronics. However, little is known about the switching effects of a single SCO molecule when it is confined between two metal electrodes. Here we examine the switching properties of a [Fe(III)(EtOSalPet )(NCS)] SCO molecule that is specifically tailored for surface deposition and binding to only one gold electrode in a nanogap device. Temperature dependent conductivity measurements on SCO molecule containing electromigrated gold break junctions show voltage independent telegraphic-like switching between two resistance states at temperature below 200 K. The transition temperature is very different from the transition temperature of 83 K that occurs in a bulk film of the same material. This indicates that the bulk, co-operative SCO phenomenon is no longer preserved for a single molecule and that the surface interaction drastically increases the temperature of the SCO phenomenon. Another key finding of this work is that some devices show switching between multiple resistance levels. We propose that in this case, two SCO molecules are present within the nanogap with both participating in the electronic transport and switching. △ Less

Submitted 19 January, 2020; originally announced January 2020.

Comments: 13 pages, 4 figures, supplementary information (5 pages, 5 figures)

Explaining Deep Classification of Time-Series Data with Learned Prototypes

Authors: Alan H. Gee, Diego Garcia-Olano, Joydeep Ghosh, David Paydarfar

Abstract: The emergence of deep learning networks raises a need for explainable AI so that users and domain experts can be confident applying them to high-risk decisions. In this paper, we leverage data from the latent space induced by deep learning models to learn stereotypical representations or "prototypes" during training to elucidate the algorithmic decision-making process. We study how leveraging prot… ▽ More The emergence of deep learning networks raises a need for explainable AI so that users and domain experts can be confident applying them to high-risk decisions. In this paper, we leverage data from the latent space induced by deep learning models to learn stereotypical representations or "prototypes" during training to elucidate the algorithmic decision-making process. We study how leveraging prototypes effect classification decisions of two dimensional time-series data in a few different settings: (1) electrocardiogram (ECG) waveforms to detect clinical bradycardia, a slowing of heart rate, in preterm infants, (2) respiration waveforms to detect apnea of prematurity, and (3) audio waveforms to classify spoken digits. We improve upon existing models by optimizing for increased prototype diversity and robustness, visualize how these prototypes in the latent space are used by the model to distinguish classes, and show that prototypes are capable of learning features on two dimensional time-series data to produce explainable insights during classification tasks. We show that the prototypes are capable of learning real-world features - bradycardia in ECG, apnea in respiration, and articulation in speech - as well as features within sub-classes. Our novel work leverages learned prototypical framework on two dimensional time-series data to produce explainable insights during classification tasks. △ Less

Submitted 4 September, 2019; v1 submitted 18 April, 2019; originally announced April 2019.

Comments: The first two authors contributed equally. Accepted May 20, Presented Jun 14, 2019 at the ICML Time-series Workshop in Long Beach, CA, USA. Accepted June 15, Presented Aug 11, 2019 at the IJCAI Workshop on Knowledge Discovery in Healthcare Data in Macao, China. Formal proceedings available in the CEUR Workshop Proceedings (http://ceur-ws.org/Vol-2429/)

Journal ref: Proceedings of the 4th International Workshop on Knowledge Discovery in Healthcare Data, co-located with the 28th International Joint Conference on Artificial Intelligence (IJCAI 2019)

Transparent Machine Education of Neural Networks for Swarm Shepherding Using Curriculum Design

Authors: Alexander Gee, Hussein Abbass

Abstract: Swarm control is a difficult problem due to the need to guide a large number of agents simultaneously. We cast the problem as a shepherding problem, similar to biological dogs guiding a group of sheep towards a goal. The shepherd needs to deal with complex and dynamic environments and make decisions in order to direct the swarm from one location to another. In this paper, we design a novel curricu… ▽ More Swarm control is a difficult problem due to the need to guide a large number of agents simultaneously. We cast the problem as a shepherding problem, similar to biological dogs guiding a group of sheep towards a goal. The shepherd needs to deal with complex and dynamic environments and make decisions in order to direct the swarm from one location to another. In this paper, we design a novel curriculum to teach an artificial intelligence empowered agent to shepherd in the presence of the large state space associated with the shepherding problem and in a transparent manner. The results show that a properly designed curriculum could indeed enhance the speed of learning and the complexity of learnt behaviours. △ Less

Submitted 3 January, 2019; originally announced March 2019.

Periodic and Aperoidic Variability in the Molecular Cloud ρOphiuchus

Authors: J. R. Parks, P. Plavchan, R. J. White, A. H. Gee

Abstract: Presented are the results of a near-IR photometric survey on 1678 stars in the direction of the ρOphiuchus (ρOph) star forming region using data from the 2MASS Calibration Database. For each target in this sample, up to 1584 individual J, H and K_s band photometric measurements with a cadence of ~1 day are obtained over 3 observing seasons spanning ~2.5 years; it is the most intensive survey of st… ▽ More Presented are the results of a near-IR photometric survey on 1678 stars in the direction of the ρOphiuchus (ρOph) star forming region using data from the 2MASS Calibration Database. For each target in this sample, up to 1584 individual J, H and K_s band photometric measurements with a cadence of ~1 day are obtained over 3 observing seasons spanning ~2.5 years; it is the most intensive survey of stars in this region to date. This survey identifies 101 variable stars with ΔK_s band amplitudes from 0.044 to 2.31 mag and Δ(J-K_s) color amplitudes ranging from 0.053 to 1.47 mag. Of the 72 young ρOph star cluster members included in this survey, 79% are variable; in addition, 22 variable stars are identified as candidate members. Based on the temporal behavior of the K_s time-series, the variability is distinguished as either periodic, long time-scale or irregular. This temporal behavior coupled with the behavior of stellar colors is used to assign a dominant variability mechanism. A new period-searching algorithm finds periodic signals in 32 variable stars with periods between 0.49 to 92 days. △ Less

Submitted 20 September, 2013; originally announced September 2013.

The Peculiar Periodic YSO WL 4 in Rho Ophiuchus

Authors: Peter Plavchan, Alan H. Gee, Karl Stapelfeldt, Andrew Becker

Abstract: We present the discovery of 130.87 day periodic near-infrared flux variability for the Class II T Tauri star WL 4 (= 2MASS J16271848-2429059, ISO-Oph 128). Our data are from the 2MASS Calibration Point Source Working Database, and constitute 1580 observations in J, H and Ks of a field in Rho Ophiuchus used to calibrate the 2MASS All-Sky Survey. We identify a light curve for WL 4 with eclipse amp… ▽ More We present the discovery of 130.87 day periodic near-infrared flux variability for the Class II T Tauri star WL 4 (= 2MASS J16271848-2429059, ISO-Oph 128). Our data are from the 2MASS Calibration Point Source Working Database, and constitute 1580 observations in J, H and Ks of a field in Rho Ophiuchus used to calibrate the 2MASS All-Sky Survey. We identify a light curve for WL 4 with eclipse amplitudes of ~0.4 mag lasting more than one-quarter the period, and color variations in J-H and H-Ks of ~0.1 mag. The long period cannot be explained by stellar rotation. We propose that WL 4 is a triple YSO system, with an inner binary orbital period of 130.87 days. We posulate that we are observing each component of the inner binary alternately being eclipsed by a circum-binary disk with respect to our line of sight. This system will be useful in investigating terrestrial zone YSO disk properties and dynamics at ~1 Myr. △ Less

Submitted 28 July, 2008; originally announced July 2008.

Comments: ApJ Letters, accepted