Search

The main technological obstacle hampering the dissemination of modern optoelectronic devices operating with large light-matter coupling strength ${\Omega}$ is an in-depth comprehension of the carrier current extraction and injection from and into strongly coupled light-matter states, the so-called polaritonic states. The main challenge lies in modeling the interaction between excitations of different nature, namely bosonic excitations (the plasmonic ISB excitations) with fermionic excitations (the electrons within the extraction or injection subband). In this work, we introduce a comprehensive quantum framework that encompasses both the ISB plasmonic mode and the extractor/injector mode, with a specific emphasis on accurately describing the coherent nature of transport. This reveals inherent selection rules dictating the interaction between the ISB plasmon and the extraction/injection subband. To incorporate the dynamics of the system, this framework is combined to a density matrix model and a quantum master equation which have the key property to distinguish intra and intersubband mechanisms. These theoretical developments are confronted to experimental photocurrent measurements from midinfrared quantum cascade detectors (${\lambda}$ = 10 ${\mu}$m) embedded in metal-semiconductor-metal microcavities, operating at the onset of the strong light-matter coupling regime (2${\Omega}$ = 9.3 meV). We are able to reproduce quantitatively the different features of the photocurrent spectra, notably the relative amplitude evolution of the polaritonic peaks with respect to the voltage bias applied to the structure. These results on extraction allow us to elucidate the possibility to effectively inject electronic excitations into ISB plasmonic states, and thus polaritonic states.

Generative adversarial networks (GANs) have many application areas including image editing, domain translation, missing data imputation, and support for creative work. However, GANs are considered 'black boxes'. Specifically, the end-users have little control over how to improve editing directions through disentanglement. Prior work focused on new GAN architectures to disentangle editing directions. Alternatively, we propose GANravel a user-driven direction disentanglement tool that complements the existing GAN architectures and allows users to improve editing directions iteratively. In two user studies with 16 participants each, GANravel users were able to disentangle directions and outperformed the state-of-the-art direction discovery baselines in disentanglement performance. In the second user study, GANravel was used in a creative task of creating dog memes and was able to create high-quality edited images and GIFs.

Generative Adversarial Network (GAN) is widely adopted in numerous application areas, such as data preprocessing, image editing, and creativity support. However, GAN's 'black box' nature prevents non-expert users from controlling what data a model generates, spawning a plethora of prior work that focused on algorithm-driven approaches to extract editing directions to control GAN. Complementarily, we propose a GANzilla: a user-driven tool that empowers a user with the classic scatter/gather technique to iteratively discover directions to meet their editing goals. In a study with 12 participants, GANzilla users were able to discover directions that (i) edited images to match provided examples (closed-ended tasks) and that (ii) met a high-level goal, e.g., making the face happier, while showing diversity across individuals (open-ended tasks).

We demonstrate mid-infrared quantum cascade detectors (QCD) operating in the strong light-matter coupling regime. They operate around $\lambda = 10~\mu m$ with a minimum Rabi splitting of 9.3 meV. A simple model based on the usual description of transport in QCDs does not reproduce the polaritonic features in the photo-current spectra. On the contrary, a more refined approach, based on the semi-classical coupled modes theory, is capable to reproduce both optical and electrical spectra with excellent agreement. By correlating absorption/photo-response with the simulations, we demonstrate that - in this system - resonant tunneling from the polaritonic states is the main extraction mechanism. The dark intersubband states are not involved in the process, contrary to what happens in electrically injected polaritonic emitters.

We examine the effect of uniform ambient magnetic fields on the evolution of supernova-driven blast waves into a homogeneous ambient ISM in thermal equilibrium. Using the Pencil Code we simulate high resolution nonideal magnetohydrodynamic simulations in 3D. We find that supernova blast waves are sensitive to plane-parallel magnetic fields of strength in excess of 1 $\mu$G for ambient gas number density 1 cm$^{-3}$ . Perpendicular to the field, the inward magnetic pressure gradient induces retrograde mass accretion in the wake of the primary shock front. Subsequently, we find that the primary shockwave expands faster perpendicular to the field, but with reduced momentum, while the remnant core is subject to magnetic confinement. This leads to a decrease in fractional volume of hot gas but also an increase in the density and temperature of hot gas in the magnetically confined remnant. The magnetic pressure gradient behind the shock front generates enhanced regions favourable to UV- heating and thus reduces net radiative losses. Although the presence of a strong uniform magnetic field can reduce momentum early on, and hence residual kinetic energy, it increases the efficiency of residual total energy injection by the SN into the ISM by up to 40% within 1 Myr.

We explore the effect of magnetic fields on the vertical distribution and multiphase structure of the supernova-driven interstellar medium (ISM) in simulations that admit dynamo action. As the magnetic field is amplified to become dynamically significant, gas becomes cooler and its distribution in the disc becomes more homogeneous. We attribute this to magnetic quenching of vertical velocity, which leads to a decrease in the cooling length of hot gas. A non-monotonic vertical distribution of the large-scale magnetic field strength, with the maximum at |z| $\approx$ 300 pc causes a downward pressure gradient below the maximum which acts against outflow driven by SN explosions, while it provides pressure support above the maximum., Comment: 10 pages, 10 figures

The role of magnetic fields in the multi-phase interstellar medium (ISM) is explored using magnetohydrodynamic (MHD) simulations that include energy injection by supernova (SN) explosions and allow for dynamo action. Apart from providing additional pressure support of the gas layer, magnetic fields reduce the density contrast between the warm and hot gas phases and quench galactic outflows. A dynamo-generated, self-consistent large-scale magnetic field affects the ISM differently from an artificially imposed, unidirectional magnetic field., Comment: 2 pages, Proceedings of FM8 "New Insights in Extragalactic Magnetic Fields", XXXth General Assembly of the IAU, Vienna, August 20-31, 2018

We apply Gaussian smoothing to obtain mean magnetic field, density, velocity, and magnetic and kinetic energy densities from our numerical model of the interstellar medium, based on three-dimensional magnetohydrodynamic equations in a shearing box $1 \times 1 \times 2 \, {\rm kpc}$ in size. The interstellar medium is highly compressible, as the turbulence is transonic or supersonic; it is thus an excellent context in which to explore the use of smoothing to represent physical variables in a compressible medium in terms of their mean and fluctuating parts. Unlike alternative averaging procedures, such as horizontal averaging, Gaussian smoothing retains the three-dimensional structure of the mean fields. Although Gaussian smoothing does not obey the Reynolds rules of averaging, physically meaningful and mathematically sound central statistical moments are defined as suggested by Germano (1992). We discuss methods to identify an optimal smoothing scale $\ell$ and the effects of this choice on the results. From spectral analysis of the magnetic, density and velocity fields, we find a suitable smoothing length for all three fields, of $\ell \approx 75 \, {\rm pc}$. Such a smoothing scale is likely to be sensitive to the choice of simulation parameters; this may be considered in future work, but here we just explore the methodology. We discuss the properties of third-order statistical moments in fluctuations of kinetic energy density in compressible flows, and suggest their physical interpretation. The mean magnetic field, amplified by a mean-field dynamo, significantly alters the distribution of kinetic energy in space and between scales, reducing the magnitude of kinetic energy at intermediate scales. This intermediate-scale kinetic energy is a useful diagnostic of the importance of SN-driven outflows., Comment: 23 pages, 10 figures. Revised, submitted to Geophysical and Astrophysical Fluid Dynamics

In this paper, we address the scheduling problem in wireless ad hoc networks by exploiting the computational advantage that comes when such scheduling problems can be represented by claw-free conflict graphs where we consider a wireless broadcast medium. It is possible to formulate a scheduling problem of network coded flows as finding maximum weighted independent set (MWIS) in the conflict graph of the network. Finding MWIS of a general graph is NP-hard leading to an NP-hard complexity of scheduling. In a claw-free conflict graph, MWIS can be found in polynomial time leading to a throughput-optimal scheduling. We show that the conflict graph of certain wireless ad hoc networks are claw-free. In order to obtain claw-free conflict graphs in general networks, we suggest introducing additional conflicts (edges) while keeping the decrease in MWIS size minimal. To this end, we introduce an iterative optimization problem to decide where to introduce edges and investigate its efficient implementation. Besides, we exemplify some physical modifications to manipulate the conflict graph of a network and also propose a mixed scheduling strategy for specific networks. We conclude that claw breaking method by adding extra edges can perform nearly optimal under the necessary assumptions., Comment: This work is to be submitted to the IEEE for possible publication

We consider a decentralized stochastic multi-armed bandit problem with multiple players. Each player aims to maximize his/her own reward by pulling an arm. The arms give rewards based on i.i.d. stochastic Bernoulli distributions. Players are not aware about the probability distributions of the arms. At the end of each turn, the players inform their neighbors about the arm he/she pulled and the reward he/she got. Neighbors of players are determined according to an Erd{\H{o}}s-R{\'e}nyi graph with connectivity $\alpha$. This graph is reproduced in the beginning of every turn with the same connectivity. When more than one player choose the same arm in a turn, we assume that only one of the players who is randomly chosen gets the reward where the others get nothing. We first start by assuming players are not aware of the collision model and offer an asymptotically optimal algorithm for $\alpha = 1$ case. Then, we extend our prior work and offer an asymptotically optimal algorithm for any connectivity but zero, assuming players aware of the collision model. We also study the effect of $\alpha$, the degree of communication between players, empirically on the cumulative regret by comparing them with traditional multi-armed bandit algorithms., Comment: This work is an extension of the paper [arXiv:1711.01628] which has been accepted to the 2017 IEEE ICMLA and submitted to Elsevier Signal Processing

In this work, we address a completely novel case which is the meeting problem of mobile nodes (players) in a wireless ad hoc network. We assume that the only information players have is the received signal strength (RSS) measurements of other players and there is no centralized communication. Each node has a different frequency band and they try to meet with each other in a decentralized manner. We consider the case where players are allowed to move in four orthogonal directions in a grid. Then, we use multi-armed bandit approach to give rewards in the end of every move that players do using RSS measurements. Taking into account that we have an adversarial setting, we propose different algorithms to solve the meeting problem. We start with considering two players and then extend the work to multiplayer by considering multiplayer problem as multiple two player problems. Throughout the paper, we construct mathematical bounds on meeting times and related concepts and evaluate the simulation results in the end., Comment: This paper has been submitted to IEEE Transactions on Signal Processing

In this paper, our goal is to compare performances of three different algorithms to predict the ratings that will be given to movies by potential users where we are given a user-movie rating matrix based on the past observations. To this end, we evaluate User-Based Collaborative Filtering, Iterative Matrix Factorization and Yehuda Koren's Integrated model using neighborhood and factorization where we use root mean square error (RMSE) as the performance evaluation metric. In short, we do not observe significant differences between performances, especially when the complexity increase is considered. We can conclude that Iterative Matrix Factorization performs fairly well despite its simplicity., Comment: This work has been accepted to the 2017 IEEE ICMLA

We consider decentralized stochastic multi-armed bandit problem with multiple players in the case of different communication probabilities between players. Each player makes a decision of pulling an arm without cooperation while aiming to maximize his or her reward but informs his or her neighbors in the end of every turn about the arm he or she pulled and the reward he or she got. Neighbors of players are determined according to an Erdos-Renyi graph with which is reproduced in the beginning of every turn. We consider i.i.d. rewards generated by a Bernoulli distribution and assume that players are unaware about the arms' probability distributions and their mean values. In case of a collision, we assume that only one of the players who is randomly chosen gets the reward where the others get zero reward. We study the effects of connectivity, the degree of communication between players, on the cumulative regret using well-known algorithms UCB1, epsilon-Greedy and Thompson Sampling., Comment: This work has been accepted to the 2017 IEEE ICMLA

We explore the effects of the multi-phase structure of the interstellar medium (ISM) on galactic magnetic fields. Basing our analysis on compressible magnetohydrodynamic (MHD) simulations of supernova-driven turbulence in the ISM, we investigate the properties of both the mean and fluctuating components of the magnetic field. We find that the mean magnetic field preferentially resides in the warm phase and is generally absent from the hot phase. The fluctuating magnetic field does not show such pronounced sensitivity to the multi-phase structure., Comment: 5 pages, 4 figures