Search

In-SRAM computing promises energy efficiency, but circuit nonlinearities and PVT variations pose major challenges in designing robust accelerators. To address this, we introduce OPTIMA, a modeling framework that aids in analyzing bit-line discharge and power consumption in 6T-SRAM-based accelerators. It provides insights into limiting factors and enables fast design-space exploration of circuit configurations. Leveraging OPTIMA for in-SRAM multiplications exhibits ~100x simulation speed-up while maintaining an RMS modeling error of 0.88mV. Exploration yields an optimized multiplier with 1.05pJ energy consumption per 4-bit operation and classification accuracies of 71.8% (top-1) and 90.4% (top-5) for ImageNet and 92.5% for CIFAR-10 datasets respectively when applied in quantized DNNs. To further support research and development, we made our tool flow available open source at https://github.com/sevjaeg/optima.

Model Predictive Control (MPC) is a successful control methodology, which is applied to increasingly complex systems. However, real-time feasibility of MPC can be challenging for complex systems, certainly when an (extremely) large number of constraints have to be adhered to. For such scenarios with a large number of state constraints, this paper proposes two novel MPC schemes for general nonlinear systems, which we call constraint-adaptive MPC. These novel schemes dynamically select at each time step a (varying) set of constraints that are included in the on-line optimization problem. Carefully selecting the included constraints can significantly reduce, as we will demonstrate, the computational complexity with often only a slight impact on the closed-loop performance. Although not all (state) constraints are imposed in the on-line optimization, the schemes still guarantee recursive feasibility and constraint satisfaction. A numerical case study illustrates the proposed MPC schemes and demonstrates the achieved computation time improvements exceeding two orders of magnitude without loss of performance., Comment: 6 pages, 4 figures, IFAC NMPC 2021 conference

Model predictive control (MPC) is an optimization-based control strategy with broad industrial adoption. Unfortunately, the required computation time to solve the receding-horizon MPC optimization problem can become prohibitively large for many applications with a large number of state constraints. This large number of state constraints can, for instance, originate from spatially discretizing a partial differential equation of which the solution has to satisfy constraints over the full spatial domain. This is particularly the case in MPC for RF-based hyperthermia cancer treatments, which forms a strong motivation for this study. To address this problem, we propose a novel constraint-adaptive MPC framework for linear discrete-time systems. In this framework, we select at each time-step a subset of the state constraints that are included in the optimization problem, thereby reducing the online computational burden. Critically, our framework guarantees the same closed-loop performance, recursive feasibility, and constraint satisfaction properties as the original (non-reduced) MPC scheme. We achieve this result by efficiently exploiting reachable set computations and the MPC cost function. We will demonstrate our novel method using a hyperthermia cancer treatment case study showing a two-orders of magnitude improvement in computation time, with identical closed-loop performance as the original (non-reduced) MPC scheme., Comment: 6 pages, 3 figures, CDC 2021 conference

The considerable body of data available for evaluating biometric recognition systems in Research and Development (R\&D) environments has contributed to the increasingly common problem of target performance mismatch. Biometric algorithms are frequently tested against data that may not reflect the real world applications they target. From a Testing and Evaluation (T\&E) standpoint, this domain mismatch causes difficulty assessing when improvements in State-of-the-Art (SOTA) research actually translate to improved applied outcomes. This problem can be addressed with thoughtful preparation of data and experimental methods to reflect specific use-cases and scenarios. To that end, this paper evaluates research solutions for identifying individuals at ranges and altitudes, which could support various application areas such as counterterrorism, protection of critical infrastructure facilities, military force protection, and border security. We address challenges including image quality issues and reliance on face recognition as the sole biometric modality. By fusing face and body features, we propose developing robust biometric systems for effective long-range identification from both the ground and steep pitch angles. Preliminary results show promising progress in whole-body recognition. This paper presents these early findings and discusses potential future directions for advancing long-range biometric identification systems based on mission-driven metrics.

Routine computed tomography (CT) scans often detect a wide range of renal cysts, some of which may be malignant. Early and precise localization of these cysts can significantly aid quantitative image analysis. Current segmentation methods, however, do not offer sufficient interpretability at the feature and pixel levels, emphasizing the necessity for an explainable framework that can detect and rectify model inaccuracies. We developed an interpretable segmentation framework and validated it on a multi-centric dataset. A Variational Autoencoder Generative Adversarial Network (VAE-GAN) was employed to learn the latent representation of 3D input patches and reconstruct input images. Modifications in the latent representation using the gradient of the segmentation model generated counterfactual explanations for varying dice similarity coefficients (DSC). Radiomics features extracted from these counterfactual images, using a ground truth cyst mask, were analyzed to determine their correlation with segmentation performance. The DSCs for the original and VAE-GAN reconstructed images for counterfactual image generation showed no significant differences. Counterfactual explanations highlighted how variations in cyst image features influence segmentation outcomes and showed model discrepancies. Radiomics features correlating positively and negatively with dice scores were identified. The uncertainty of the predicted segmentation masks was estimated using posterior sampling of the weight space. The combination of counterfactual explanations and uncertainty maps provided a deeper understanding of the image features within the segmented renal cysts that lead to high uncertainty. The proposed segmentation framework not only achieved high segmentation accuracy but also increased interpretability regarding how image features impact segmentation performance.

Large Language Models (LLMs) have demonstrated great performance in few-shot In-Context Learning (ICL) for a variety of generative and discriminative chemical design tasks. The newly expanded context windows of LLMs can further improve ICL capabilities for molecular inverse design and lead optimization. To take full advantage of these capabilities we developed a new semi-supervised learning method that overcomes the lack of experimental data available for many-shot ICL. Our approach involves iterative inclusion of LLM generated molecules with high predicted performance, along with experimental data. We further integrated our method in a multi-modal LLM which allows for the interactive modification of generated molecular structures using text instructions. As we show, the new method greatly improves upon existing ICL methods for molecular design while being accessible and easy to use for scientists.

The interplay between densification and positional ordering during the process of crystal nucleation is a greatly investigated topic. Even for the simplest colloidal model -- hard spheres -- there has been much debate regarding the potential foreshadowing of nucleation by significant fluctuations in either local density or local structure. Considering anisotropic particles instead of spheres adds a third degree of freedom to the self-organization process of crystal nucleation: orientational ordering. Here, we investigate the crystal nucleation of hard triangular prisms. Using Monte Carlo simulations, we first carefully determine the crystal-fluid coexistence values and calculate the nucleation barriers for two degrees of supersaturation. Next, we use brute force simulations to obtain a large set of spontaneous nucleation events. By studying the time evolution of the local density, positional ordering, and orientational ordering in the region in which the nucleus first arises, we demonstrate that all local order parameters increase simultaneously from the very start of the nucleation process. We thus conclude that we observe no precursor for the crystal nucleation of hard triangular prisms.

In the study of crystal nucleation via computer simulations, hard spheres are arguably the most extensively explored model system. Nonetheless, even in this simple model system, the complex thermodynamics of crystal nuclei can sometimes give rise to counterintuitive results, such as the recent observation that the pressure inside a critical nucleus is lower than that of the surrounding fluid, seemingly clashing with the strictly positive Young--Laplace pressure we would expect in liquid droplets. Here, we re-derive many of the founding equations associated with crystal nucleation, and use the hard-sphere model to demonstrate how they give rise to this negative pressure difference. We exploit the fact that, in the canonical ensemble, a nucleus can be in a (meta)stable equilibrium with the fluid, and measure the surface stress for both flat and curved interfaces. Additionally, we explain the effect of defects on the chemical potential inside the crystal nucleus. Lastly, we present a simple, fitted thermodynamic model to capture the properties of the nucleus, including the work required to form critical nuclei.

This paper reports the design and experimental characterization of a cryogenic compact low-power 60GHz amplifier for control of electron/hole spin qubits, as elementary building block for monolithic Si quantum processors. Tested at 2 K, the amplifier exhibits S21 of 15 dB at 59 GHz, BW3dB of 52.5-67.5 GHz, and power consumption of 2.16 mW. Owing to the topology with inductorless active network, the amplifier has a compact core area of 0.18 x 0.19 mm2., Comment: The final version of this draft has been appearing in the proceedings of IMS 2022 available on IEEE Xplore Digital Library (10.1109/IMS37962.2022.9865561)

This paper reports the experimental characterization at the cryogenic temperature of a compact mm-wave broadband single-pole single-throw (SPST) switch in 22nm FDSOI CMOS technology. The switch consists of two n-MOSFETs with a special device option to reduce the substrate parasitic effects, and a third n-MOSFET to improve isolation. Unlike prior wideband mm-wave switches, it does not require any large passive components, allowing a very compact design, low loss and high isolation performance. The cryogenic measurements at 2 K show an insertion loss lower than 2.3 dB, an isolation better than 25.3 dB, and the return loss better than -11.5 dB, over the entire frequency range from DC to 70 GHz., Comment: The final version of this draft has been appearing in the Proceedings of IMS 2022 available on IEEE Xplore Digital Library (10.1109/IMS37962.2022.9865577)

The formation of protein precursors, due to the condensation of atomic carbon under the low-temperature conditions of the molecular phases of the interstellar medium, opens alternative pathways for the origin of life. We perform peptide synthesis under conditions prevailing in space and provide a comprehensive analytic characterization of its products. The application of 13C allowed us to confirm the suggested pathway of peptide formation that proceeds due to the polymerization of aminoketene molecules that are formed in the C + CO + NH3 reaction. Here, we address the question of how the efficiency of peptide production is modified by the presence of water molecules. We demonstrate that although water slightly reduces the efficiency of polymerization of aminoketene, it does not prevent the formation of peptides.

One method for computationally determining phase boundaries is to explicitly simulate a direct coexistence between the two phases of interest. Although this approach works very well for fluid-fluid coexistences, it is often considered to be less useful for fluid-crystal transitions, as additional care must be taken to prevent the simulation boundaries from imposing unwanted strains on the crystal phase. Here, we present a simple adaptation to the direct coexistence method that nonetheless allows us to obtain highly accurate predictions of fluid-crystal coexistence conditions. We test our approach on hard spheres, the screened Coulomb potential, and a 2D patchy-particle model. In all cases, we find excellent agreement between the direct coexistence approach and (much more cumbersome) free-energy calculation methods. Moreover, the method is sufficiently accurate to resolve the (tiny) free-energy difference between the face-centered cubic and hexagonally close-packed crystal of hard spheres in the thermodynamic limit. The simplicity of this method also ensures that it can be trivially implemented in essentially any simulation method or package. Hence, this approach provides an excellent alternative to free-energy based methods for the precise determination of phase boundaries., Comment: Consists of a manuscript, SI, and sample LAMMPS script

To ensure efficient and reliable operation of a concentrating solar-thermal power (CSP) plant, its solar collector field needs to accurately focus sunlight. The optical efficiency and structural integrity of the solar collectors is significantly influenced by wind conditions in the field. In this study, we present insights into dynamic wind loading on parabolic trough CSP collectors. We derive novel conclusions by analyzing a first-of-a-kind measurement campaign of wind and structural loads, performed at an operational CSP plant. Previous research primarily relied on wind tunnel tests and simulations, leaving uncertainty about wind loading effects in operational settings. We demonstrate that the parabolic trough field significantly alters the turbulent wind field within the collector field, especially under winds perpendicular to the trough rows. Our measurements within the trough field show reduced wind speeds, changes in wind direction and turbulence properties, and vortex shedding from the trough assemblies. These modifications to the wind field directly impact both static and dynamic support structure loads. Our measurements reveal higher wind loads on trough assemblies compared to those observed previously in wind tunnel tests. The insights from this study offer a novel perspective on our understanding of wind-driven loads on CSP collectors. By informing the development of next-generation design tools and models, this research paves the way for enhanced structural integrity and improved optical performance in future parabolic trough systems., Comment: 33 pages, 9 figures

With the constant spread of misinformation on social media networks, a need has arisen to continuously assess the veracity of digital content. This need has inspired numerous research efforts on the development of misinformation detection (MD) models. However, many models do not use all information available to them and existing research contains a lack of relevant datasets to train the models, specifically within the South African social media environment. The aim of this paper is to investigate the transferability of knowledge of a MD model between different contextual environments. This research contributes a multimodal MD model capable of functioning in the South African social media environment, as well as introduces a South African misinformation dataset. The model makes use of multiple sources of information for misinformation detection, namely: textual and visual elements. It uses bidirectional encoder representations from transformers (BERT) as the textual encoder and a residual network (ResNet) as the visual encoder. The model is trained and evaluated on the Fakeddit dataset and a South African misinformation dataset. Results show that using South African samples in the training of the model increases model performance, in a South African contextual environment, and that a multimodal model retains significantly more knowledge than both the textual and visual unimodal models. Our study suggests that the performance of a misinformation detection model is influenced by the cultural nuances of its operating environment and multimodal models assist in the transferability of knowledge between different contextual environments. Therefore, local data should be incorporated into the training process of a misinformation detection model in order to optimize model performance., Comment: Artificial Intelligence Research. SACAIR 2023

The current best practice approach for the retrospective diagnosis of adverse drug events (ADEs) in hospitalized patients relies on a full patient chart review and a formal causality assessment by multiple medical experts. This evaluation serves to qualitatively estimate the probability of causation (PC); the probability that a drug was a necessary cause of an adverse event. This practice is manual, resource intensive and prone to human biases, and may thus benefit from data-driven decision support. Here, we pioneer a causal modeling approach using observational data to estimate a lower bound of the PC (PC$_{low}$). This method includes two key causal inference components: (1) the target trial emulation framework and (2) estimation of individualized treatment effects using machine learning. We apply our method to the clinically relevant use-case of vancomycin-induced acute kidney injury in intensive care patients, and compare our causal model-based PC$_{low}$ estimates to qualitative estimates of the PC provided by a medical expert. Important limitations and potential improvements are discussed, and we conclude that future improved causal models could provide essential data-driven support for medication safety monitoring in hospitalized patients., Comment: Extended Abstract presented at Machine Learning for Health (ML4H) symposium 2023, December 10th, 2023, New Orleans, United States, 14 pages

Generating single photons on demand in silicon is a challenge to the scalability of silicon-on-insulator integrated quantum photonic chips. While several defects acting as artificial atoms have recently demonstrated an ability to generate antibunched single photons, practical applications require tailoring of their emission through quantum cavity effects. In this work, we perform cavity quantum electrodynamics experiments with ensembles of artificial atoms embedded in silicon-on-insulator microresonators. The emitters under study, known as W color centers, are silicon tri-interstitial defects created upon self-ion implantation and thermal annealing. The resonators consist of circular Bragg grating cavities, designed for moderate Purcell enhancement ($F_p=12.5$) and efficient luminescence extraction ($\eta_{coll}=40\%$ for a numerical aperture of 0.26) for W centers located at the mode antinode. When the resonant frequency mode of the cavity is tuned with the zero-phonon transition of the emitters at 1218 nm, we observe a 20-fold enhancement of the zero-phonon line intensity, together with a two-fold decrease of the total relaxation time in time-resolved photoluminescence experiments. Based on finite-difference time-domain simulations, we propose a detailed theoretical analysis of Purcell enhancement for an ensemble of W centers, considering the overlap between the emitters and the resonant cavity mode. We obtain a good agreement with our experimental results assuming a quantum efficiency of $65 \pm 10 \%$ for the emitters in bulk silicon. Therefore, W centers open promising perspectives for the development of on-demand sources of single photons, harnessing cavity quantum electrodynamics in silicon photonic chips.

(Abridged) We investigate the impact of radiative feedback from massive stars on their natal cloud and focus on the transition from the HII region to the atomic PDR (crossing the ionisation front (IF)), and the subsequent transition to the molecular PDR (crossing the dissociation front (DF)). We use high-resolution near-IR integral field spectroscopic data from NIRSpec on JWST to observe the Orion Bar PDR as part of the PDRs4All JWST Early Release Science Program. The NIRSpec data reveal a forest of lines including, but not limited to, HeI, HI, and CI recombination lines, ionic lines, OI and NI fluorescence lines, Aromatic Infrared Bands (AIBs including aromatic CH, aliphatic CH, and their CD counterparts), CO2 ice, pure rotational and ro-vibrational lines from H2, and ro-vibrational lines HD, CO, and CH+, most of them detected for the first time towards a PDR. Their spatial distribution resolves the H and He ionisation structure in the Huygens region, gives insight into the geometry of the Bar, and confirms the large-scale stratification of PDRs. We observe numerous smaller scale structures whose typical size decreases with distance from Ori C and IR lines from CI, if solely arising from radiative recombination and cascade, reveal very high gas temperatures consistent with the hot irradiated surface of small-scale dense clumps deep inside the PDR. The H2 lines reveal multiple, prominent filaments which exhibit different characteristics. This leaves the impression of a "terraced" transition from the predominantly atomic surface region to the CO-rich molecular zone deeper in. This study showcases the discovery space created by JWST to further our understanding of the impact radiation from young stars has on their natal molecular cloud and proto-planetary disk, which touches on star- and planet formation as well as galaxy evolution., Comment: 52 pages, 30 figures, submitted to A&A