Search

This roadmap consolidates recent advances while exploring emerging applications, reflecting the remarkable diversity of hardware platforms, neuromorphic concepts, and implementation philosophies reported in the field. It emphasizes the critical role of cross-disciplinary collaboration in this rapidly evolving field.

We present a reformulation of QM/MM as a fully quantum mechanical theory of interacting subsystems, all treated at the level of density functional theory (DFT). For the MM subsystem, which lacks orbitals, we assign an ad hoc electron density and apply orbital-free DFT functionals to describe its quantum properties. The interaction between the QM and MM subsystems is also treated using orbital-free density functionals, accounting for Coulomb interactions, exchange, correlation, and Pauli repulsion. Consistency across QM and MM subsystems is ensured by employing data-driven, many-body MM force fields that faithfully represent DFT functionals. Applications to water-solvated systems demonstrate that this approach achieves unprecedented, very rapid convergence to chemical accuracy as the size of the QM subsystem increases. We validate the method with several pilot studies, including water bulk, water clusters (prism hexamer and pentamers), solvated glucose, a palladium aqua ion, and a wet monolayer of MoS$_2$.

Matrix-vector multiplications (MVMs) are essential for a wide range of applications, particularly in modern machine learning and quantum computing. In photonics, there is growing interest in developing architectures capable of performing linear operations with high speed, low latency, and minimal loss. Traditional interferometric photonic architectures, such as the Clements design, have been extensively used for MVM operations. However, as these architectures scale, improving stability and robustness becomes critical. In this paper, we introduce a novel photonic braid interferometer architecture that outperforms both the Clements and Fldzhyan designs in these aspects. Using numerical simulations, we evaluate the performance of these architectures under ideal conditions and systematically introduce non-idealities such as insertion losses, beam splitter imbalances, and crosstalk. The results demonstrate that the braid architecture offers superior robustness due to its symmetrical design and reduced layer count. Further analysis shows that the braid architecture is particularly advantageous in large-scale implementations, delivering better performance as the size of the interferometer increases. We also assess the footprint and total insertion losses of each architecture. Although waveguide crossings in the braid architecture slightly increase the footprint and insertion loss, recent advances in crossing technology significantly minimize these effects. Our study suggests that the braid architecture is a robust solution for photonic neuromorphic computing, maintaining high fidelity in realistic conditions where imperfections are inevitable.

We propose a novel hybrid mode interferometer (HMI) leveraging the interference of hybridized TE-TM modes in a silicon-on-insulator (SOI) waveguide integrated with a GeSe phase change material (PCM) layer. The SOI waveguide's dimensions are optimized to support the hybridization of the fundamental transverse magnetic ($TM_0$) and the first higher transverse electric ($TE_1$) mode. This design allows for efficient and nearly equal power coupling between these two modes, resulting in high-contrast interference when starting from the amorphous PCM state. The PCM's phase transition induces a differential change in the modal effective index, enabling high-contrast transmittance modulation. Our numerical simulations demonstrate a multilevel transmission with a high contrast of nearly 14 dB, when the amorphous region's length is varied incrementally, enabling multi-bit storage. The transmittance is maximized in the fully crystalline state with an insertion loss below 0.1 dB. The HMI can also operate as a quasi-pure phase shifter when partially amorphized, making it suitable for Mach-Zehnder interferometers. These characteristics make the proposed device a promising candidate for applications in photonic memories and neuromorphic computing.

The kinetic energy (KE) kernel, which is defined as the second order functional derivative of the KE functional with respect to density, is the key ingredient to the construction of KE models for orbital free density functional theory (OFDFT) applications. For solids, the KE kernel is usually approximated using the uniform electron gas (UEG) model or the UEG-with-gap model. These kernels do not have information about the effects from the core electrons since there are no orbitals for the projection on nonlocal pseudopotentials. To illuminate this aspect, we provide a methodology for computing the KE kernel from Kohn-Sham DFT and apply it to the valence electrons in bulk aluminum (Al) with a face-centered cubic lattice and in bulk silicon (Si) in a semiconducting crystal diamond state. We find that bulk-derived local pseudopotentials provide accurate results for the KE kernel in the interstitial region. The effect of using nonlocal pseudopotentials manifests at short wavelengths, defined by the diameter of an ion surrounded by its core electrons. Specifically, we find that the utilization of nonlocal pseudopotentials leads to significant deviations in the KE kernel from the von Weizsacker result in this region, which is, as a rule, explicitly enforced in most widely used KE functional approximations for OFDFT simulations.

In this work, we discuss our vision for neuromorphic accelerators based on integrated photonics within the framework of the Horizon Europe NEUROPULS project. Augmented integrated photonic architectures that leverage phase-change and III-V materials for optical computing will be presented. A CMOS-compatible platform will be discussed that integrates these materials to fabricate photonic neuromorphic architectures, along with a gem5-based simulation platform to model accelerator operation once it is interfaced with a RISC-V processor. This simulation platform enables accurate system-level accelerator modeling and benchmarking in terms of key metrics such as speed, energy consumption, and footprint., Comment: 4 pages, 3 figures

Photonic Neural Networks (PNNs) are gaining significant interest in the research community due to their potential for high parallelization, low latency, and energy efficiency. PNNs compute using light, which leads to several differences in implementation when compared to electronics, such as the need to represent input features in the photonic domain before feeding them into the network. In this encoding process, it is common to combine multiple features into a single input to reduce the number of inputs and associated devices, leading to smaller and more energy-efficient PNNs. Although this alters the network's handling of input data, its impact on PNNs remains understudied. This paper addresses this open question, investigating the effect of commonly used encoding strategies that combine features on the performance and learning capabilities of PNNs. Here, using the concept of feature importance, we develop a mathematical methodology for analyzing feature combination. Through this methodology, we demonstrate that encoding multiple features together in a single input determines their relative importance, thus limiting the network's ability to learn from the data. Given some prior knowledge of the data, however, this can also be leveraged for higher accuracy. By selecting an optimal encoding method, we achieve up to a 12.3% improvement in accuracy of PNNs trained on the Iris dataset compared to other encoding techniques, surpassing the performance of networks where features are not combined. These findings highlight the importance of carefully choosing the encoding to the accuracy and decision-making strategies of PNNs, particularly in size or power constrained applications.

The optimization of fuel-optimal low-thrust collision avoidance maneuvers (CAMs) in scenarios involving multiple encounters between spacecraft is addressed. The optimization's objective is the minimization of the total fuel consumption while respecting constraints on the total probability of collision. The solution methodology combines sequential convex programming, second-order cone programming, and differential algebra to approximate the non-convex optimal control problem progressively. A Gaussian mixture model method is used to propagate the initial covariance matrix of the secondary spacecraft, allowing us to split it into multiple mixands that can be treated as different objects. This leads to an accurate propagation of the uncertainties. No theoretical guarantee is given for the convergence of the method to the global optimum of the original optimal control problem. Nonetheless, good performance is demonstrated through case studies involving multiple short- and long-term encounters, showcasing the generation of fuel-efficient CAMs while respecting operational constraints., Comment: Submitted to the Journal of Guidance Control and Dynamics

A simple and reliable algorithm for collision avoidance maneuvers (CAMs), capable of computing impulsive, multi-impulsive, and low-thrust maneuvers, is proposed. The probability of collision (PoC) is approximated by a polynomial of arbitrary order as a function of the control, transforming the CAM designinto a polynomial program. The solution procedure is initiated by computing the CAM via a first-order greedy optimization approach, wherein the control action is applied in the direction of the gradient of PoC to maximize its change. Successively, the polynomial is truncated at higher orders, and the solution of the previous order is used to linearize the constraint. This enables achieving accurate solutions even for highly nonlinear safety metrics and dynamics. Since the optimization process comprises only polynomial evaluations, the method is computationally efficient, with run times typically below 1 s. Moreover, no restrictions on the considered dynamics are necessary; therefore, results are shown for Keplerian, J2, and circular restricted three-body problem dynamics., Comment: Submitted to IEEE Transaction on Aerospace and Electronic Systems

The study of regularity in signals can be of great importance, typically in medicine to analyse electrocardiogram (ECG) or electromyography (EMG) signals, but also in climate studies, finance or security. In this work we focus on security primitives such as Physical Unclonable Functions (PUFs) or Pseudo-Random Number Generators (PRNGs). Such primitives must have a high level of complexity or entropy in their responses to guarantee enough security for their applications. There are several ways of assessing the complexity of their responses, especially in the binary domain. With the development of analog PUFs such as optical (photonic) PUFs, it would be useful to be able to assess their complexity in the analog domain when designing them, for example, before converting analog signals into binary. In this numerical study, we decided to explore the potential of the disentropy of autocorrelation as a measure of complexity for security primitives as PUFs, TRNGs or PRNGs with analog output or responses. We compare this metric to others used to assess regularities in analog signals such as Approximate Entropy (ApEn) and Fuzzy Entropy (FuzEn). We show that the disentropy of autocorrelation is able to differentiate between well-known PRNGs and non-optimised or bad PRNGs in the analog and binary domain with a better contrast than ApEn and FuzEn. Then, we show that the disentropy of autocorrelation is able to detect small patterns injected in PUFs responses., Comment: 11 pages, 7 figures

In this work we introduce a novel subsystem-based electronic structure embedding method that combines the projection-based block-orthogonalized Manby-Miller embedding (BOMME) with the density-based Frozen Density Embedding (FDE) methods. Our approach is effective for systems in which the building blocks interact at varying strengths while still maintaining a lower computational cost compared to a quantum simulation of the entire system. To evaluate the performance of our method, we assess its ability to reproduce the X-ray absorption spectra (XAS) of chloride and fluoride anions in aqueous solutions (based on a 50-water droplet model) via real-time time-dependent density functional theory (rt-TDDFT) calculations. We employ an ensemble approach to compute XAS for the K- and L-edges, utilizing multiple snapshots of configuration space obtained from classical molecular dynamics simulations with a polarizable force field. Configurational averaging influences both the broadening of spectral features and their intensities, with contributions to the final intensities originating from different geometry configurations. We found that embedding models that are too approximate for halide-water specific interactions, as in the case of FDE, fail to reproduce the experimental spectrum for chloride. Meanwhile, BOMME tends to overestimate intensities, particularly for higher energy features because of finite-size effects. Combining FDE for the second solvation shell and retaining BOMME for the first solvation shell mitigates this effect, resulting in an overall improved agreement within the energy range of the experimental spectrum. Additionally, we compute the transition densities of the relevant transitions, confirming that these transitions occur within the halide systems. Thus, our real-time QM/QM/QM embedding method proves to be a promising approach for modeling XAS of solvated systems.

In the contemporary security landscape, the incorporation of photonics has emerged as a transformative force, unlocking a spectrum of possibilities to enhance the resilience and effectiveness of security primitives. This integration represents more than a mere technological augmentation; it signifies a paradigm shift towards innovative approaches capable of delivering security primitives with key properties for low-power systems. This not only augments the robustness of security frameworks, but also paves the way for novel strategies that adapt to the evolving challenges of the digital age. This paper discusses the security layers and related services that will be developed, modeled, and evaluated within the Horizon Europe NEUROPULS project. These layers will exploit novel implementations for security primitives based on physical unclonable functions (PUFs) using integrated photonics technology. Their objective is to provide a series of services to support the secure operation of a neuromorphic photonic accelerator for edge computing applications., Comment: 6 pages, 4 figures

Hydrogen at extreme temperatures and pressures is ubiquitous throughout our universe and naturally occurs in a variety of astrophysical objects. In addition, it is of key relevance for cutting-edge technological applications, with inertial confinement fusion research being a prime example. In the present work, we present exact \emph{ab initio} path integral Monte Carlo (PIMC) results for the electronic density of warm dense hydrogen along a line of constant degeneracy across a broad range of densities. Using the well-known concept of reduced density gradients, we develop a new framework to identify the breaking of bound states due to pressure ionization in bulk hydrogen. Moreover, we use our PIMC results as a reference to rigorously assess the accuracy of a variety of exchange--correlation (XC) functionals in density functional theory calculations for different density regions. Here a key finding is the importance of thermal XC effects for the accurate description of density gradients in high-energy density systems. Our exact PIMC test set is freely available online and can be used to guide the development of new methodologies for the simulation of warm dense matter and beyond.

This work presents a sequential convex program method to compute fuel-optimal collision avoidance maneuvers for long-term encounters. The low-thrust acceleration model is used to account for the control, but the method can compute high-thrust maneuvers by increasing the maximum available acceleration. Dealing with the long-term conjunction poses additional challenges compared to the short-term problem because the encounter is not instantaneous. Thus, under the assumption of Gaussian statistics, the probability of collision is replaced by a simpler metric, the instantaneous probability of collision (IPoC) and a keep-out zone constraint is formulated as a continuous condition to be respected throughout the time frame of interest. The robustness of the solution is improved by introducing a constraint on the sensitivity of the IPoC. Furthermore, the collision avoidance problem is coupled with the classical station-keeping requirement for geostationary Earth orbit satellites and with a return to the nominal orbit condition for low Earth orbit satellites. Even though no guarantee is given for the recovery of the global optimum solution, numerical simulations in different orbital regimes show that the proposed approach can yield a local fuel-optimal solution with a run-time suitable for autonomous applications., Comment: Submitted to Journal of Guidance Control and Dynamics (first revision)

This special session paper introduces the Horizon Europe NEUROPULS project, which targets the development of secure and energy-efficient RISC-V interfaced neuromorphic accelerators using augmented silicon photonics technology. Our approach aims to develop an augmented silicon photonics platform, an FPGA-powered RISC-V-connected computing platform, and a complete simulation platform to demonstrate the neuromorphic accelerator capabilities. In particular, their main advantages and limitations will be addressed concerning the underpinning technology for each platform. Then, we will discuss three targeted use cases for edge-computing applications: Global National Satellite System (GNSS) anti-jamming, autonomous driving, and anomaly detection in edge devices. Finally, we will address the reliability and security aspects of the stand-alone accelerator implementation and the project use cases., Comment: 10 pages, 2 figures, conference

The field of neuromorphic computing has been rapidly evolving in recent years, with an increasing focus on hardware design and reliability. This special session paper provides an overview of the recent developments in neuromorphic computing, focusing on hardware design and reliability. We first review the traditional CMOS-based approaches to neuromorphic hardware design and identify the challenges related to scalability, latency, and power consumption. We then investigate alternative approaches based on emerging technologies, specifically integrated photonics approaches within the NEUROPULS project. Finally, we examine the impact of device variability and aging on the reliability of neuromorphic hardware and present techniques for mitigating these effects. This review is intended to serve as a valuable resource for researchers and practitioners in neuromorphic computing., Comment: 10 pages, 4 figures, 4 tables

Orbital-free density functional theory (OF-DFT) constitutes a computationally highly effective tool for modeling electronic structures of systems ranging from room-temperature materials to warm dense matter. Its accuracy critically depends on the employed kinetic energy (KE) density functional, which has to be supplied as an external input. In this work we consider several nonlocal and Laplacian-level KE functionals and use an external harmonic perturbation to compute the static density response at T=0 K in the linear and beyond linear response regimes. We test for the satisfaction of exact conditions in the limit of uniform densities and for how approximate KE functionals reproduce the density response of realistic materials (e.g., Al and Si) against the Kohn-Sham DFT reference which employs the exact KE. The results illustrate that several functionals violate exact conditions in the UEG limit. We find a strong correlation between the accuracy of the KE functionals in the UEG limit and in the strongly inhomogeneous case. This empirically demonstrates the importance of imposing the limit of UEG response for uniform densities and validates the use of the Lindhard function in the formulation of kernels for nonlocal functionals. This conclusion is substantiated by additional calculations for bulk Aluminum (Al) with a face-centered cubic (fcc) lattice and Silicon (Si) with an fcc lattice, body-centered cubic (bcc) lattice, and semiconducting crystal diamond (cd) state. The analysis of fcc Al, and fcc as well as bcc Si data follows closely the conclusions drawn for the UEG, allowing us to extend our conclusions to realistic systems that are subject to density inhomogeneities induced by ions.

The theorems of density functional theory (DFT) and reduced density matrix functional theory (RDMFT) establish a bijective map between the external potential of a many-body system and its electron density or one-particle reduced density matrix. Building on this foundation, we show that machine learning can be used to generate surrogate electronic structure methods. In particular, we generate surrogates of local and hybrid DFT as well as Hartree-Fock and full configuration interaction theories for systems ranging from small molecules such as water to more complex compounds like propanol and benzene. The surrogate models use the one-electron reduced density matrix as the central quantity to be learned. From the predicted density matrices, we show that either standard quantum chemistry or a second machine-learning model can be used to compute molecular observables, energies, and atomic forces. We show that the surrogate models can generate essentially anything that a standard electronic structure method can, ranging from band gaps and Kohn-Sham orbitals to energy-conserving ab-initio molecular dynamics simulations and IR spectra, which account anharmonicity and thermal effects, without the need of computationally expensive algorithms such as self-consistent field theory. The algorithms developed here are packaged in an efficient and easy to use Python code, QMLearn, accessible by the broader community on popular platforms.

We present a new methodology for the linear-response time-dependent density functional theory (LR-TDDFT) calculation of the dynamic density response function of warm dense matter in an adiabatic approximation that can be used with any available exchange-correlation (XC) functional across Jacob's Ladder and across temperature regimes. The main novelty of the presented approach is that it can go beyond the adiabatic local density approximation (ALDA) and generalized LDA (AGGA) while preserving the self-consistence between the Kohn-Sham (KS) response function and adiabatic XC kernel for extended systems. The key ingredient for the presented method is the combination of the adiabatic XC kernel from the direct perturbation approach with the macroscopic dynamic KS response from the standard LR-TDDFT method using KS orbitals. We demonstrate the application of the method for the example of warm dense hydrogen, for which we perform a detailed analysis of the KS density response function, the RPA result, the total density response function and of the adiabatic XC kernel. The analysis is performed using LDA, GGA, and meta-GGA level approximations for the XC effects. The presented method is directly applicable to disordered systems such as liquid metals, warm dense matter, and dense plasmas.