HAL-INSA Toulouse
Not a member yet
34325 research outputs found
Sort by
Unconventional Scalings of Quantum Entropies in Long-Range Heisenberg Chains
No full textInternational audienceIn this work, building on state-of-the-art quantum Monte Carlo simulations, we perform systematic finite-size scaling of both entanglement and participation entropies for long-range Heisenberg chain with unfrustrated power-law decaying interactions. We find distinctive scaling behaviors for both quantum entropies in the various regimes explored by tuning the decay exponent , thus capturing non-trivial features through logarithmic terms, beyond the case of linear Nambu-Goldstone modes. Our systematic analysis reveals that the quantum entanglement information, hidden in the scaling of the two studied entropies, can be obtained to the same level of order parameters and other usual finite-size observables of quantum many-body lattice models. The analysis and results obtained here can readily apply to more quantum criticalities in 1D and 2D systems
Verifying Properties of Binary Neural Networks Using Sparse Polynomial Optimization
No full text22 pages, 2 figures, 7 tablesInternational audienceThis paper explores methods for verifying the properties of Binary Neural Networks (BNNs), focusing on robustness against adversarial attacks. Despite their lower computational and memory needs, BNNs, like their full-precision counterparts, are also sensitive to input perturbations. Established methods for solving this problem are predominantly based on Satisfiability Modulo Theories and Mixed-Integer Linear Programming techniques, which are characterized by NP complexity and often face scalability issues. We introduce an alternative approach using Semidefinite Programming relaxations derived from sparse Polynomial Optimization. Our approach, compatible with continuous input space, not only mitigates numerical issues associated with floating-point calculations but also enhances verification scalability through the strategic use of tighter first-order semidefinite relaxations. We demonstrate the effectiveness of our method in verifying robustness against both and -based adversarial attacks
Photoinduced modulation of the oxidation state of dibenzothiophene S-oxide molecules on an insulating substrate
No full textInternational audienceOn-surface chemistry aims to overcome the limitations of conventional in-solution synthesis by taking advantage of the confinement in two dimensions to master highly ordered covalent structures with tailored properties. So far, most of the reported work is conducted on metal substrates and relies on unconventional mechanisms, thereby precluding a direct transposition of well-established organic reactions from solutions to surfaces. In addition, the intrinsic properties and reactivity of metal substrates often limit the activation methods available to trigger on-surface reactions, and photoinduced processes are especially difficult to handle due to quenching of the adsorbed precursor molecules. Herein, the photoinduced deoxygenation of dibenzothiophene S -oxide (DBTO) derivatives is transposed from solutions to insulating alkali halide surfaces in ultra-high vacuum. By combining in-solution and on-surface investigations by means of scanning tunneling microscopy, non-contact atomic force microscopy, as well as bias spectroscopy measurements, we provide evidence of the successful on-surface deoxygenation of individual DBTO derivatives under UV irradiation. The photoinduced deoxygenation is conducted at low temperature (<25 K) on a NaCl thin film formed on a Au(111) substrate to yield the reduced dibenzothiophene (DBT) product with excellent chemoselectivity. This work thus opens the way to in-situ photocontrolled charge state manipulation in purely organic compounds
Convective self-aggregation as a two-regime damped gravity wave
No full textConvective self-aggregation is the spontaneous spatial organisation of deep-convective clouds into a limited region surrounded by drier, convectively-inhibited regions, which occurs in many models. We propose a simple, piecewise linear model for self-aggregation based on primitive equations. In this model, each atmospheric column is in one of two possible thermodynamic regimes, deep-convective or convectively inhibited, and the thermodynamics in each regime is linearised. The model simulates aggregated and non-aggregated steady states and reproduces many properties of self-aggregation as simulated by kilometre-resolution models. In particular, it exhibits an hysteresis with multiple, aggregated and non-aggregated equilibria and a similar sensitivity to convective inhibition, domain size, and boundary-layer radiative cooling in the dry region as in kilometre-resolution simulations. These results suggest that a self-aggregated state can be considered as a simple gravity wave with two phases: one convective and one convectively inhibited
Minimax density estimation in the adversarial framework under local differential privacy
No full textWe consider the problem of nonparametric density estimation under privacy constraints in an adversarial framework. To this end, we study minimax rates over Sobolev spaces under local differential privacy. We first obtain a lower bound which allows us to quantify the impact of privacy compared with the classical framework. Next, we introduce a new Coordinate block privacy mechanism that guarantees local differential privacy, which, coupled with a projection estimator, achieves the minimax optimal rates. Finally, we develop an adaptive procedure which is optimal in the minimax sense up to logarithmic terms
Metabolism of procyanidins by the human gut microbiota: Impact of their interaction with cell wall and modifications induced by processing
No full textInternational audienc
Efficient Computation of Laser Self-Mixing Sensor Signal under Variable Optical Feedback
No full textInternational audienc
Cascaded Voltage-Current Control for Grid-Forming Inverters: Design of Multi-Resonant State-Feedback Controller Using LMI Approach
No full textInternational audienceGrid-forming (GFM) inverters play a critical role in regulating voltage and frequency to ensure stable operation of isolated microgrids. Among the various control strategies, resonant controllers are recognized as one of the highestperformance solutions for AC current and voltage control. Traditionally, state feedback control implementations based on single loop control have been widely used for GFM inverters. However, these structures exhibit limitations in addressing key issues such as current reference tracking, and overcurrent conditions. To overcome these challenges, this paper proposes a discrete state feedback proportional multi-resonant control strategy implemented within a cascaded voltage-current loop for GFM inverters. This approach enables the controller to handle both current and voltage reference tracking simultaneously. A set of linear matrix inequality (LMI) constraints is employed to synthesize controller gains ensuring robust stability. Simulation results demonstrate the effectiveness of the proposed control strategy in achieving harmonic mitigation and precise reference tracking
Conventional micro and nanofabrication techniques and processes
No full textInternational audienceThis presentation will show how established micro- and nanofabrication techniques contribute to progress in electronics, sensing technologies, and medical applications. After introducing the cleanroom environment, essential for controlling contamination and ensuring high-precision fabrication, it will highlight versatile processes for creating thin films, patterning materials, and defining structures at micro- and nanoscales.The talk will showcase advanced lithography and etching techniques that enable the miniaturization of components, unlocking new possibilities for high-performance devices. Examples of ongoing projects, such as the development of sensitive gas sensors for environmental and health applications, will illustrate the transition from fabrication processes to real-world solutions.Finally, an overview of current research will demonstrate how micro- and nanotechnologies are evolving to meet the demands of future applications, particularly in life sciences and cancer diagnostics, paving the way for smarter, more efficient, and more integrated devices
Doubly ionised 2,4-imidazolidinedione (hydantoin) dissociations induced by electron collision, fragmentation thresholds vs vertical excitation energies
No full textInternational audienceThe double ionisation/dissociation process for a prebiotic molecule (hydantoin) induced by 20-40 eV electron collisions was studied using time-of-flight spectrometry and velocity map imaging. Fragmentation into two molecular cations reveals that hydrogen transfer occurs in the excited state. Furthermore, the thresholds of the main dissociation channels were measured. Comparison with state-of-the-art calculations of the vertical excited states of hydantoin2+ gives an accurate indication of the initial population of the excited states.</div