Welcome to the website of the annual Symposium. This year’s theme will be 'Cognitive Matters - Physics of cognitive advancements'. During this event speakers will be talking about their research revolving around cognitive materials and systems.
Materials form the basis of our environment and is a big focus point in the field of applied physics. Development of new kinds of materials have lead to big technological advancements of things like 3D printing, smart materials, perovskite solar cells. In recent years there has been a lot of research in smart and self-learning materials. This year's symposium will have the research field of cognitive systems and materials as its focus point. This field of research comprises the fields of (applied) physics, mathematics, computer science and artificial intelligence. Our symposium will mainly focus on the (applied) physics. Six speakers will share their knowledge on new developments in different sub-fields of cognitive systems and materials.
As the organization we hope to welcome you the 12th of May to, what we hope, is a compelling and interesting day.
If you live in Groningen and sign up before 8th of may, free lunch from Broodje Ben will be delivered to your house. You can choose between either beef, salmon or beet sandwich. An email will be send to you where you can provide your choice and address.
Prof. dr. Beatriz Noheda
Dr. ir. Yoeri van de Burgt
Dr. Julie Grollier
Dr. Siegfried Karg
Prof. Dr. ir. Wilfred van der Wiel
Dr. Javier del Valle
Dr. Laura Begon-Lours
Brain-inspired or neuromorphic computing could address the inherent limitations of conventional silicon technology in dedicated machine learning applications. Organic electronic materials have shown potential to overcome some of these limitations. This talk describes state-of-the-art organic neuromorphic devices and provides an overview of the current challenges in the field and attempts to address them. I demonstrate a novel concept based on an organic electrochemical transistor and show how crucial challenges in the field such as stability, variability and linearity can be overcome. Next to that, organic electronic materials have the potential to operate at the interface with biology. This can pave the way for novel architectures with bio-inspired features, offering promising solutions for the manipulation and the processing of biological signals and potential applications ranging from brain-computer-interfaces and smart robotics to bioinformatics. I will highlight our recent efforts for such hybrid biological memory devices.
In this talk, I want to show that the joint development of new materials, nanodevices and algorithms is essential for the development of smart and environmentally friendly chips, which inherently learn from their physics. The goal of neuromorphic computing is to take inspiration from the architecture of the brain to build a low-power circuit for artificial intelligence. I will first give a brief overview of the state of the art of neuromorphic computing, highlighting the opportunities offered by emerging nanodevices in this field, and the related challenges. I will then show how physics can be exploited at the algorithm and device level to solve these challenges and unlock learning on a chip. I will illustrate the need for this material-algorithm co-development through examples of our work on an algorithm called Equilibrium Propagation that minimizes both error and energy of a dynamic system, and through the implementation of spintronic neural networks that communicate through radio-frequency waves
Bio-inspired hardware holds the promise of intelligent, low-energy and highly reconfigurable computing schemes, apt to tackle complex computational problems as pattern matching and image recognition. Recent advances in computing methods based on oscillatory dynamical systems motivated the research for Beyond CMOS material and devices. New materials such as the transition-metal oxide VO₂ hold physical characteristics which can be efficiently employed to perform computations. In this talk, we will discuss the implementation of oscillatory neural networks based on the near-room temperature insulator-to-metal transition of VO₂. Self-sustained oscillations at the voltage output of the VO₂ nanoscale devices, caused by the continuous switching of the VO₂ between its insulating and metallic phase, can be achieved through electrical activation near the phase transition threshold. Therefore, arrays of compact relaxation oscillators can be fabricated. The oscillators are locked in frequency through coupling with simple electrical components. The phase-relations of the frequency- locked oscillators are used to store and process information, in particular for the case applications of pattern recognition and acceleration of convolutional neural networks.
The strong increase in digital computing power in combination with the availability of large amounts of data has led to a revolution in Machine Learning. Computers now exhibit superhuman performance in activities such as pattern recognition and board games. However, the implementation of Machine Learning in digital computers is intrinsically wasteful, with energy consumption becoming prohibitively high for many applications. For that reason, people have started looking at natural information processing systems, in particular the brain, that operate much more efficiently. Whereas the brain specifically utilizes the biochemistry and electrophysics of wet, soft biological tissue for information processing, one could in principle exploit any material substrate and its physical properties to solve a problem. In this presentation, we give examples of how nanomaterial networks can be trained using the principle of Material Learning to take full advantage of the computational power of matter . We have shown that a designless network of gold nanoparticles can be configured into Boolean logic gates using artificial evolution . We demonstrate that this principle is generic and can be transferred to other material systems as well. By exploiting the nonlinearity of a nanoscale network of boron dopants in silicon (Si:B networks), we can significantly facilitate (nonlinear) classification. Using a convolutional neural network approach, it becomes possible to use our device for handwritten digit recognition . An alternative Material Learning is approach is followed by first mapping our Si:B network on a deep neural network model, which allows for applying standard Machine Learning techniques in finding functionality . Finally, we show that the widely applied Machine Learning technique of gradient descent can be directly applied in materio, opening up the pathway for autonomously learning hardware systems .
Neuromorphic computing, which mimics the architecture and components of biological neural networks, is an emerging technology which might overcome some of the challenges that traditional computing is facing. A neuromorphic computer is composed of two basic elements: neurons and synapses. We show how resistive switching can be used to mimic the functionalities of these two elements, focusing on volatile switching caused by the voltage-triggered insulator to metal transition in VO₂ and V₂O₃. We will discuss several aspects of this transition, such as the underlying mechanism, the dynamics and its spatial distribution. We will also show some of our recent efforts towards emulating neuronal behaviour using this phenomenology.
Deep-Neural-Networks accelerators aim at implementing the Vector-Matrix-Multiplication in the analog domain, by a parallel voltage drop through a cross-bar array of tuneable resistances, the memristors. By analogy with the brain, these elements, connecting the different layers of neurons, are the “synapses” of the neural network. Among the different technologies available for the fabrication of memristors, ferroelectric-based devices show analog resistive switching, excellent retention properties and small stochasticity. However the fabrication of ferroelectric memristors was for a long time incompatible with industrial processes. With the discovery of a ferroelectric phase in Hafnia (a material already widely used in fabrication) in 2011, the ferroelectric based devices receive an increasing interest. In this talk, we will discuss the development of ferroelectric memristors from an industry perspective, in particular how we constrain our processes to CMOS compatible technologies. Two technologies will be presented, the two and three terminals ferroelectric memristors.
These days, a small USB stick costing only €10 can hold up to 16 GB of data. In hospitals, a camera the size of a pill can be swallowed to survey a patient's intestines. Modern pacemakers, critical devices that control abnormal heart rhythms, are now less than a tenth the size of earlier ones. And in the oceans, tiny GPS transmitters track endangered turtles to help protect them.
While these devices are incredibly small, they represent a big milestone in technological progress. At the heart of each of these life-enhancing innovations is a microchip a tiny package of integrated circuitry that powers the performance of the device.
In a world in which major breakthroughs measure only a few nanometres in size, the constant quest is to produce chips that are smaller, faster, more effective and less expensive. One of the major high-tech players leading the quest is ASML, a manufacturer of lithography systems for producing computer chips.
Founded in 2002, Bright Spark first emerged due to its inventions in the fields of water treatment. But Bright Spark offers more than that. Bright Spark is a continuous inventor in all kinds of areas. This is because a creative mind does not detract from problems or regulations but is constantly looking for new challenges.
Bright Spark reconciles creativity, technical knowledge and corporate drive to develop durable, effective and cost saving water treatment systems for the international market and is proud on having developed a unique technique to disinfect water. The system works due to electrolysis and kills germs and viruses in drinking water and is successfully applied in many different ways.
The company is known for its creative developments in all possible (technical) fields. Next to water treatment systems, Bright Spark also develops sensors, pedal canoes and since recently also technical art objects. Bright Spark is part of the knowledge network in the north of The Netherlands and works closely together with Wetsus, a technical top institute for water. Therefore, Bright Spark is constantly involved in fundamental and innovative projects that can lead to creative solutions!
From left to right: