Les jonctions tunnel magnétiques sont des candidats prometteurs for le calcul. Mais quand elles sont réduites à des dimensions nanométriques, conserver leur stabilité devient difficile. Les jonctions tunnel magnétiques instables subissent des renversements aléatoires de leur aimantation et se comportent comme des oscillateurs stochastiques. Pourtant, la nature stochastique de ces jonctions tunnel superparamagnétiques n’est pas une faille mais un atout qui peut être utilisé pour le calcul bio-inspiré. En effet, notre cerveau a évolué de sorte qu’il puisse fonctionner dans un environnement bruité et avec des composants instables. Dans cette thèse, nous montrons plusieurs applications possibles des jonctions tunnel superparamagnétiques.Nous démontrons qu’une junction tunnel superparamagnétique peut être synchronisée en fréquence et en phase à une faible tension oscillante. De manière contre intuitive, notre expérience montre que cela peut être fait grâce à l’injection de bruit dans le système. Nous développons un modèle théorique pour comprendre ce phénomène et prédire qu’il permet un gain énergétique d’un facteur cent par rapport à la synchronisation d’oscillateurs à transfert de spin traditionnels. De plus, nous utilisons notre modèle pour étudier la synchronisation de plusieurs jonctions couplées. De nombreuses méthodes théoriques réalisant des tâches cognitives telles que la reconnaissance de motifs et la classification grâce à la synchronisation d’oscillateurs ont été proposés. Utiliser la synchronisation induite par le bruit de jonctions tunnel superparamagnétiques permettrait de réaliser ces tâches à basse énergie.Nous faisons une analogie entre les jonctions tunnel superparamagnétiques et les neurones sensoriels qui émettent des pics de tension séparés par des intervalles aléatoires. En poursuivant cette analogie, nous démontrons que des populations de jonctions tunnel superparamagnétiques peuvent représenter des distributions de probabilité et réaliser de l’inférence Bayésienne. De plus, nous démontrons que des populations interconnectées peuvent faire du calcul, notamment de l’apprentissage, des transformations de coordonnées et de la fusion sensorielles. Un tel système est faisable de manière réaliste et pourrait permettre de fabriquer des capteurs intelligents à bas coût énergétique. / Magnetic tunnel junctions are promising candidates for computing applications. But when they are reduced to nanoscale dimensions, maintaining their stability becomes an issue. Unstable magnetic tunnel junctions undergo random switches of the magnetization between their two stable states and thus behave as stochastic oscillators. However, the stochastic nature of these superparamagnetic tunnel junctions is not a liability but an asset which can be used for the implementation of bio-inspired computing schemes. Indeed, our brain has evolved to function in a noisy environment and with unstable components. In this thesis, we show several possible applications of superparamagnetic tunnel junctions.We demonstrate how a superparamagnetic tunnel junction can be frequency and phase-locked to a weak oscillating voltage. Counterintuitively, our experiment shows that this is achieved by injecting noise in the system. We develop a theoretical model to understand this phenomenon and predict that it allows a hundred-fold energy gain over the synchronization of traditional dc-driven spin torque oscillators. Furthermore, we leverage our model to study the synchronization of several coupled junctions. Many theoretical schemes using the synchronization of oscillators to perform cognitive tasks such as pattern recognition and classification have been proposed. Using the noise-induced synchronization of superparamagnetic tunnel junctions would allow implementing these tasks at low energy.We draw an analogy between superparamagnetic tunnel junctions and sensory neurons which fire voltage pulses with random time intervals. Pushing this analogy, we demonstrate that populations of junctions can represent probability distributions and perform Bayesian inference. Furthermore, we demonstrate that interconnected populations can perform computing tasks such as learning, coordinate transformations and sensory fusion. Such a system is realistically implementable and could allow for intelligent sensory processing at low energy cost.

Avec l'avènement de l'"intelligence artificielle", les ordinateurs, appareils mobiles et objets connectés sont amenés à dépasser les calculs arithmétiques et logiques pour lesquels ils ont été optimisés durant des décennies, afin d'effectuer des tâches "cognitives" telles que la traduction automatique ou la reconnaissance d'images et de voix, et pour lesquelles ils ne sont pas adaptés. Ainsi, un super-calculateur peut-il consommer des mégawatts pour effectuer des tâches que le cerveau humain traite avec 20 watt. Par conséquent, des système de calcul alternatifs inspirés du cerveau font l'objet de recherches importantes. En particulier, les oscillations neurales semblant être liées à certains traitements de données dans le cerveau ont inspiré des approches détournant la physique complexe des réseaux d'oscillateurs couplés pour effectuer des tâches cognitives efficacement. Cette thèse se fonde sur les avancées récentes en nano-technologies permettant la fabrication de nano-oscillateurs hautement intégrables pour proposer et étudier de nouvelles architectures neuro-inspirées de classification de motifs exploitant la dynamique des oscillateurs couplés et pouvant être implémentées sur puce. / With the advent of "artificial intelligence", computers, mobile devices and other connected objects are being pushed beyond the realm of arithmetic and logic operations, for which they have been optimized over decades, in order to process "cognitive" tasks such as automatic translation and image or voice recognition, for which they are not the ideal substrate. As a result, supercomputers may require megawatts to process tasks for which the human brain only needs 20 watt. This has revived interest into the design of alternative computing schemes inspired by the brain. In particular, neural oscillations that appear to be linked to computational activity in the brain have inspired approaches leveraging the complex physics of networks of coupled oscillators in order to process cognitive tasks efficiently. In the light of recent advances in nano-technology allowing the fabrication of highly integrable nano-oscillators, this thesis proposes and studies novel neuro-inspired oscillator-based pattern classification architectures that could be implemented on chip.

Dans la prochaine ère de l'informatique distribuée, les ordinateurs inspirés par le cerveau qui effectuent des opérations localement plutôt que dans des serveurs distants seraient un avantage majeur en réduisant les coûts énergétiques et réduisant l'impact environnemental. Une nouvelle génération de nanodispositifs de mémoire non-volatile est un candidat de premier plan pour réaliser cette vision neuromorphique. À l'aide de travaux théoriques et expérimentaux, nous avons exploré les problèmes critiques qui se posent lors de la réalisation physique des architectures de réseaux de neurones artificiels modernes (ANN) en utilisant des dispositifs de mémoire émergents (nanodispositifs « memristifs »). Dans notre travail expérimental, nos dispositifs organiques (polymeriques) se sont adaptés avec succès et automatiquement en tant que portes logiques reconfigurables en coopérant avec un neurone digital et programmable (FGPA). Dans nos travaux théoriques, nous aussi avons considéré les multicouches memristives ANNs. Nous avons développé et simulé des variantes de projection aléatoire (un système NoProp) et de rétropropagation (un système perceptron multicouche) qui utilisent deux crossbars. Ces systèmes d'apprentissage locaux ont montré des dépendances critiques sur les contraintes physiques des nanodispositifs. Enfin, nous avons examiné comment les conceptions ANNs “feed-forward” peuvent être modi-fiées pour exploiter les effets temporels. Nous avons amélioré la bio-inspiration et la performance du système NoProp, par exemple, avec des effets de plasticité dans la première couche. Ces effets ont été obtenus en utilisant un nanodispositif à ionisation d'argent avec un comportement de transition de plasticité intrinsèque. / In the next era of distributed computing, brain-based computers that perform operations locally rather than in remote servers would be a major benefit in reducing global energy costs. A new generation of emerging nonvolatile memory devices is a leading candidate for achieving this neuromorphic vision. Using theoretical and experimental work, we have explored critical issues that arise when physically realizing modern artificial neural network (ANN) architectures using emerging memory devices (“memristors”). In our experimental work, we showed organic nanosynapses adapting automatically as logic gates via a companion digital neuron and programmable logic cell (FGPA). In our theoretical work, we also considered multilayer memristive ANNs. We have developed and simulated random projection (NoProp) and backpropagation (Multilayer Perceptron) variants that use two crossbars. These local learning systems showed critical dependencies on the physical constraints of nanodevices. Finally, we examined how feed-forward ANN designs can be modified to exploit temporal effects. We focused in particular on improving bio-inspiration and performance of the NoProp system, for example, we improved the performance with plasticity effects in the first layer. These effects were obtained using a silver ionic nanodevice with intrinsic plasticity transition behavior.

In 2027 the Large Hadron Collider (LHC) at CERN will enter a high luminosity phase, deliver- ing 3000 fb 1 over the course of ten years. The High Luminosity LHC (HL-LHC) will increase the instantaneous luminosity delivered by a factor of 5 compared to the current operation pe- riod. This will impose significant technical challenges on all aspects of the ATLAS detector but particularly the Inner Detector, trigger, and data acquisition systems. In addition, many of the components of the Inner Detector are reaching the end of their designed lifetime and will need to be exchanged. As such, the Inner Detector will be entirely replaced by an all silicon tracker, known as the Inner Tracker (ITk). The layout of the Pixel and strip detectors will be optimised for the upgrade and will extend their forward coverage. To reduce the per-pixel hit rate and explore novel techniques for deal- ing with the conditions in HL-LHC, an inter-experiment collaboration called RD53 has been formed. RD53 is tasked with producing a front-end readout chip to be used as part of hybrid Pixel detectors that can deal with the high multiplicity environment in the HL-LHC. A silicon sensor, which makes up the other half of the hybrid Pixel detector, must also be designed to cope with the high fluences in HL-LHC. Significant damage will be caused by non- ionising energy loss in the sensor over its lifetime. This damage must be incorporated into the detector simulation both to predict the detector performance at specific conditions and to understand the e↵ects of radiation damage on data taking. The implementation of radiation damage in the ATLAS simulation framework is discussed in this thesis. Collisions produced by the HL-LHC also presents a challenge for the current track reconstruc- tion software. High luminosity is obtained, in part, by increasing the number of interactions per bunch crossing, which in turn increases the time taken for track reconstruction. Various ap- proaches to circumvent the strain on projected resources are being explored, including porting existing algorithms to parallel architectures. A popular algorithm used in track reconstruction, the Kalman filter, has been implemented in a neuromorphic architecture: IBM’s TrueNorth. The limits of using such an architecture for tracking, as well as how its performance compares to a non-spiking Kalman filter implementation, are explored in this thesis.

How can the brain be such a good computer? Part of the answer lies in the astonishing number of neurons and synapses that process electrical impulses in parallel. Part of it must be found in the ability of the nervous system to evolve in response to external stimuli and grow, sharpen, and depress synaptic connections. However, we are far from understanding even the basic mechanisms that allow us to think, be aware, recognize patterns, and imagine. The brain can do all this while consuming only around 20 Watts, out-competing any human-made processor in terms of energy-efficiency. This question is of particular interest in a historical era and technological stage where phrases like machine learning and artificial intelligence are more and more widespread, thanks to recent advances produced in the field of computer science. However, brain-inspired computation is today still relying on algorithms that run on traditional silicon-made, digital processors. Instead, the making of brain-like hardware, where the substrate itself can be used for computation and it can dynamically update its electrical pathways, is still challenging. In this work, I tried to employ organic semiconductors that work in electrolytic solutions, called organic mixed ionic-electronic conductors (OMIECs) to build hardware capable of computation. Moreover, by exploiting an electropolymerization technique, I could form conducting connections in response to electrical spikes, in analogy to how synapses evolve when the neuron fires. After demonstrating artificial synapses as a potential building block for neuromorphic chips, I shifted my attention to the implementation of such synapses in fully operational networks. In doing so, I borrowed the mathematical framework of a machine learning approach known as reservoir computing, which allows computation with random (neural) networks. I capitalized my work on demonstrating the possibility of using such networks in-vivo for the recognition and classification of dangerous and healthy heartbeats. This is the first demonstration of machine learning carried out in a biological environment with a biocompatible substrate. The implications of this technology are straightforward: a constant monitoring of biological signals and fluids accompanied by an active recognition of the presence of malign patterns may lead to a timely, targeted and early diagnosis of potentially mortal conditions. Finally, in the attempt to simulate the random neural networks, I faced difficulties in the modeling of the devices with the state-of-the-art approach. Therefore, I tried to explore a new way to describe OMIECs and OMIECs-based devices, starting from thermodynamic axioms. The results of this model shine a light on the mechanism behind the operation of the organic electrochemical transistors, revealing the importance of the entropy of mixing and suggesting new pathways for device optimization for targeted applications.

No description available.

No description available.

No description available.

No description available.

No description available.