Information processing is vital for living systems and involves complex networks of active processes. These systems have influenced various forms of modern machine learning, including reservoir computing. Reservoir computing utilizes networks of nodes with fading memory to perform computations and make complex predictions. Reservoirs can be implemented on computer hardware or unconventional physical substrates like mechanical oscillators, spins, or bacteria, known as physical reservoir computing.
We demonstrate physical reservoir computing with a synthetic active microparticle system that self-organizes from an active and passive component into inherently noisy nonlinear dynamical units. The self-organization and dynamical response of the unit is the result of a delayed propulsion of the microswimmer to a passive target. A reservoir of such units with a self-coupling via the delayed response can perform predictive tasks despite the strong noise resulting from the Brownian motion of the microswimmers. To achieve efficient noise suppression, we introduce an architecture that uses historical reservoir states for output. We discuss the node and collective reservoir dynamics.
We investigate the optical properties of opto-thermally assembled reconfigurable three-dimensional photonic crystals through localized optical heating of a thin gold film on a glass substrate. The assembly process is aided by the resulting thermal gradient induced hydrodynamic, thermophoretic as well as depletion effects. The band structure and the corresponding stop bands of the photonic crystals are probed using Fourier plane imaging and angle resolved spectroscopy of locally excited dye molecules present in the fluidic solution. The results hold direct implications for low power manipulation and assembly of functional photonic structures.
We show assembly of functional and reconfigurable three-dimensional photonic crystals aided by opto-thermal effects due to localized optical heating of a thin gold film on a glass substrate. The optical stop bands of the photonic crystals are probed using Fourier plane imaging and angle resolved spectroscopy of locally excited dye molecules present in the solution. Additionally, dark field scattering spectroscopy indicates the structural colors of the assembled structures and changes with the lattice constants. The results have direct implications for low power manipulation and assembly of functional photonic structures.
Microfluidics is commonly ruled by pressure driven flows enabling the transport of material on large scales incorporating different kinds of functionality for sensing flow control or chemical synthesis. Yet, a local control of fluids and dissolved species is difficult due to the macroscopic nature of the exerted pressure gradients.
Here we present our efforts to control liquids and dissolved species at the microscale using thermo-fluidic approaches. We employ optically controlled thermo-osmotic, thermophoretic, and thermoviscous flows to induce fluid flow to sense, localise, or separate different species in solution. We introduce different spectroscopic and microscopic signals to report on the local properties and composition of the solution with the help of machine learning approaches to track and classify species in real time to provide a feedback to steer the system into desired directions.
Reservoir computing as a highly efficient architecture for recurrent neural networks has been implemented in a variety of ways, including anharmonic oscillators, liquid surfaces, and optical and electronic circuits.
Here, we investigate whether active particle networks that mimic fundamental dynamical processes of living systems can serve as reservoirs. In particular, we realize active particle oscillators, each consisting of an immobile and an active colloidal microparticle suspended in a layer of a liquid solution. The motion of the active particles is manipulated by a feedback system using a focused laser that stimulates the particles to float in 2D by thermophoresis [1]. The active particle is programmed to be attracted to the immobile particle with a delayed response that exhibits a pitchfork bifurcation, which introduces nonlinearity and memory into the response of a single active oscillator.
Using time multiplexing, the propulsion of the oscillator is selected at different times as virtual nodes of a reservoir that are coupled to an input layer. Since the motion of the active particle is affected in a nonlinear manner with a memory of its previous state, the last node state is naturally coupled to the other nodes from different iterations due to the intrinsic property of the delayed oscillator. We illustrate the performance of the reservoir consisting of multiple oscillators with different delays by the tasks of nonlinear prediction and classification.
[1]. F. Martin et al. ACS Nano 15, 2, 3434-3440 (2021)
Artificial microswimmers are active particles designed to mimic the behavior of living microorganisms. The adaptive behavior of the latter is based on the experience they gain through interactions with the environment. They are also subjected to Brownian motion at these length scales which randomizes their position and propulsion direction making it a key feature in the adaptation process. However, artificial systems are limited in their ability to adapt to such noise and environmental stimuli. In this work, we combine artificial microswimmers with a reinforcement learning algorithm to explore their adaptive behavior in a noisy environment. These self-thermophoretic active particles are propelled and steered by generating thermal gradients on their surface with a tightly focused laser beam. They are also imaged under a microscope in real-time to monitor their dynamics. With such a versatile platform capable of real-time control and monitoring, we demonstrated the solution to a standard navigation problem under the inevitable influence of Brownian motion by introducing deep reinforcement learning, specifically deep-Q-learning. We also identified different noises in the system and how they affected the learning speed and navigation strategies picked up by the microswimmer.
The manipulation of micro- and nano-objects is of great technological significance to construct new materials, manipulate tiny amounts of liquids in fluidic systems, or detect minute concentrations of analytes. It is commonly approached by the generation of potential energy landscapes, for example, with optical fields.
Here we show that strong hydrodynamic boundary flows enable the trapping and manipulation of nano-objects near surfaces. These thermo-osmotic flows are induced by modulating the van der Waals interaction at a solid-liquid interface with optically induced temperature fields. We use a thin gold film on a glass substrate to provide localized but reconfigurable point-like optical heating. Convergent boundary flows with velocities of tens of micrometres per second are observed and substantiated by a quantitative physical model. The hydrodynamic forces acting on suspended nanoparticles and attractive van der Waals or depletion induced forces enable precise positioning and guiding of the nanoparticles. Fast multiplexing of flow fields further provides the means for parallel manipulation of many nano-objects. Our findings have direct consequences for the field of plasmonic nano-tweezers as well as other thermo-plasmonic trapping schemes and pave the way for a general scheme of nanoscopic manipulation with boundary flows.
[1] Fränzl, M. & Cichos, F. Hydrodynamic manipulation of nano-objects by optically induced thermo-osmotic flows. Nat Commun 13, 656 (2022).
We introduce a toolbox to calibrate microscopic force fields by analyzing the trajectories of a Brownian particle using machine learning, namely, recurrent neural networks. We demonstrate that this machine-learning approach outperforms standard methods when characterizing the force fields generated by harmonic potentials if the available data are limited. More importantly, it provides a tool to calibrate force fields in situations for which there are no standard methods, such as non-conservative and time-varying force fields. In order to make this method readily available for other users, we provide a Python software package named DeepCalib, which can be easily personalized and optimized for specific force fields and applications.
There is a very limited number of methods to analyze experimental trajectories of systems with feedback and time delay. In most cases, an analytical approach is not even possible. In this study, we show that the feedback parameters and the delay can be accurately characterized using machine learning, namely recurrent neural networks. We demonstrate that our method can dramatically expand the number of time-delayed feedback scenarios that we can characterize. We exemplify our findings on different numerical and experimental scenarios.
Amyloid fibrils are highly stable and organized peptide or protein structures that on the one hand can cause partially severe diseases such as Alzheimer's disease and on the other hand play fundamental roles during a plethora of biological processes. Nevertheless, there are still plenty open questions concerning their formation. We present a thermophoretic trap which is able to confine the Brownian motion of single amyloid fibrils via temperature gradients. The time-resolved tracking of the fibrils' rotational diffusion coefficients in presence of monomers permits to extract their growth rates or to directly observe secondary growth processes as fragmentation.
The calibration of physical force fields from particle trajectories is important for experiments in soft matter, biophysics, active matter, and colloidal science. However, it is not always possible to have a standard method to characterize a force field, especially for systems that are out of equilibrium. Here, we introduce a generic toolbox for calibrating any kind of conservative or non-conservative, fixed or time-varying potentials that is powered by recurrent neural networks (RNN). We show that with the help of neural networks, we can outperform standard methods as well as analyze systems that cannot be approached by existing methods. We provide a software package that is available online for free access.
We present an adapted single shot neural network architecture (YOLO) for the real-time localization and classification of particles in optical microscopy. Our work is aimed at the manipulation of microscopic objects in real-time by a feedback loop. The network is implemented in Python/Keras using the TensorFlow backend. The trained model is then exported to a GPU supported C library for real-time inference readily integrable in other programming languages such as C++ and LabVIEW. It is capable of localizing and classifying several hundred of microscopic objects even at very low signal-to-noise ratios running for images as large as 416 x 416 pixels with an inference time of about 10 ms.
We demonstrate real-time detection in tracking and manipulating active particles of different types. Symmetric active particles, as well as Janus particles propelled by self-thermophoretic laser-induced processes, are identified and controlled via a Photon-Nudging procedure developed in the group.
Artificial microswimmers are designed to mimic the self-propulsion of microscopic living organisms to yield access to the complex behavior of active matter. As compared to their living counterparts, they have only limited ability to adapt to environmental signals or retain a physical memory. Yet, different from macroscopic living systems and robots, both microscopic living organisms and artificial microswimmers are subject to thermal noise as a key feature in microscopic systems.
Here we combine real-world artificial active particles with machine learning algorithms to explore their adaptive behavior in a noisy environment with reinforcement learning. We use a real-time control of self-thermophoretic active particles to demonstrate the solution of a standard navigation problem with single and multiple swimmers and show that noise decreases the learning speed, increases the decision strength and modifies the optimal behavior based on a delayed response in the noisy environment.
The generation of localized temperature gradients is accompanied by new fundamental physics and also provides new tools for the control of molecules, particles or more complex matter in solution. We describe experiments, which use metal nano- and microstructures as optically pumped heat sources. Heat flowing from these structures along solid/liquid interfaces sets liquids into motion. With the help of such thermo-osmotic creep flows, we can trap particles and single molecules suspended in liquids without body forces but with forces balances. Also, the compression of macro-molecules becomes accessible. The inhomogeneous temperature, however, also modifies the Brownian dynamics. We report applications in the field polymer physics and protein aggregation, where such trapping techniques provide a unique new insight. We address the dynamics of heated colloids in optical tweezers with nanosecond time resolution and picometer spatial resolution to understand thermal non-equilibrium effects.
We demonstrate the long time trapping of single DNA molecules in liquids by feedback driven dynamic temperature fields. By spatially and temporally varying the temperature at a plasmonic nanostructure, thermophoretic drifts are induced that are used to trap single nano-objects. A feedback controlled switching of local temperature fields allows us to confine the motion of a single DNA molecule for minutes. The DNA conformation and conformation dynamics are analyzed in terms of a principle component analysis. Current results are in agreement with previous measurements in thermal equilibrium and suggest only a weak influence of the inhomogeneous temperature rise on the structure and dynamics in the trap.
We study the motion of a Janus particle in an inhomogeneous external temperature field generated by an optically heated gold nanoparticle. The Janus particle consists of a polystyrene particle covered on one hemisphere with a 50 nm gold film. The Janus particle is held in the vicinity of the immobilized gold nanoparticle by photon nudging, which actively propels the Janus particle towards a target. Close to the heat source, the propulsion is switched off. We find an angle dependent repulsion of the particle from the heat source. Further, an angular velocity of the Janus particle is measured, which results in an active polarization of the Janus particle in the temperature field.
We study numerically the measurement of distances and distance fluctuations by photothermal correlation spectroscopy
and coupled plasmon resonances. Gold nanoparticle dimers form a coupled longitudinal plasmon resonance
in the absorption cross section, which strongly depends on distance. This new plasmon resonance can be
advantageously used to heat the particles in a photothermal microscope. We calculate the distance dependence
of the photothermal signal as a function of particle size and distance. The results demonstrate that the photothermal
signal autocorrelation function stay single exponential even for large amplitude fluctuations and thus
directly reveals the dynamics of the distance fluctuations without any corrections as required for fluorescence
resonance energy transfer (FRET). Further, we show, that this type of distance detection provides distance
measures beyond the accessible range of a few nanometers as in FRET.
We study the fluorescence intermittency of individual Dil-molecules on silicon dioxide surfaces with the focus on intermittency statistics on timescales above 15 milliseconds. On these time-scales intermittency statistics is no longer dominated by triplet blinking. We show that rather broad distributions of dark states must be present which give rise to power law distributions for on- and off-times. The off- time distribution depends on excitation intensity.
The directed surface passivation of semiconductor CdSe, 0r CdSe/ZnS quantum dots (QD) by meso-pyridyl substituted porphyrins (H2P) has been realized via a reversible non-covalent self-assembly interaction of H2P meso-pyridyl nitrogens with ions of the ZnS shell or Cd atoms of the CdSe core in various solvents at ambient temperature. The formation of "QD-porphyrin" nanoassemblies leads to a QD photoluminescence (PL) quenching (intensity decrease and PL decay shortening) accompanied by a H2P fluorescence enhancement. The analysis of experimental Foerster resonance energy transfer efficiencies EFRET (FRET) found via acceptor (H2P) sensibilization and donor (QD) PL quenching shows that EFRET values obtained from fluorescence enhancement are of the order of 6 - 8 % for most QD studied and are thus much smaller as compared to the PL quenching efficiency. With respect to QD PL quenching efficiencies, smaller values of EFRET might be due to different competing reasons: the presence of two independent quenching processes in the nanoassemblies, energy transfer QD -> H2P and photoinduced (electron/hole) charge transfer (CT) or time-dependent QD interface dynamics leading to a noticeable QD PL quenching. The analysis of spectroscopic and kinetic findings reveals that a limited number of "vacancies" accessible for porphyrin attachment is available on the QD surface. Simultaneous presence of porphyrin triads/pentads and QDs in a solution leads to the formation of higly organzed nanoassemblies.
The cathode spot formation within first 22 ns was investigated by laser absorption photography and ps-pulse interferometry. The discharge was initiated between W-, Ag-, AuNi-, Pd- electrodes with cathode-anode distance below 100 micrometers , the arc duration was some milliseconds and arc current 5 A. A ps-pulse holographic interferometer and momentary absorption photography enabled us to determinate spatial-temporal density distributions in the ignition phase of the cathode spot. An absolute electron density value of order of 3 - 1026 m-3 has been determined indicating high conductivity values of the metal vapor plasma. Present measurements show that cathode spot plasma is essentially non- ideal and verify theoretical calculations resulting in an ionization potential decrease in dense cathode plasmas.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.