The supersymmetric contributions to the muon anomalous magnetic moment a_μ and to the decay μ → eγ are given by very similar Feynman diagrams. Previous works reported correlations in specific scenarios, in particular if a_μ is dominated by a single diagram. In this work we give an extensive survey of the possible correlations. We discuss examples of single-diagram domination with particularly strong correlations, and provide corresponding benchmark parameter points. We show how the correlations are weakened by significant cancellations between diagrams in large parts of the MSSM parameter space. Nevertheless, the order of magnitude of BR(μ → eγ) for a fixed flavor-violating parameter can often be predicted. We summarize the behavior by plotting the correlations as well as resulting bounds on the flavor-violating parameters under various assumptions on the MSSM spectrum.

Deep learning algorithms are widely used for pattern recognition in electronic noses, which are sensor arrays for gas mixtures. One of the challenges of using electronic noses is sensor drift, which can degrade the accuracy of the system over time, even if it is initially trained to accurately estimate concentrations from sensor data. In this paper, an effective drift compensation method is introduced that adds sensor drift information during training of a neural network that estimates gas concentrations. This is achieved by concatenating a calibration feature vector with sensor data and using this as an input to the neural network. The calibration feature vector is generated via a masked-autoencoder-based feature extractor trained with transfer samples, and acts as a prompt to convey sensor drift information. Our method is tested on a 3-year gas sensor array drift dataset, showing that a neural network using our method performs better than other models, including a network with additional fine tuning, demonstrating that our method is efficient at compensating for sensor drift. In this study, the effectiveness of using prompts for network training is confirmed, which better compensates for drifts in new sensor signals than network fine-tuning.

Target detection and classification by Raman spectroscopy are important techniques for biological and chemical defense in military operations. Conventionally, these techniques preprocess the observed spectra using smoothing or baseline correction and apply detection algorithms like the generalized likelihood ratio test, independent component analysis, nonnegative matrix factorization, etc. These conventional detection algorithms need preprocessing and multiple shots of Raman spectra to get a reasonable accuracy. Recently, techniques based on deep learning are being used for target detection and classification due to its great adaptability and high accuracy over other methods and due to no requirement for preprocessing. Deep learning may give a good performance, but need retraining when untrained class targets are introduced which is time-consuming and bothersome. We devise a novel algorithm using a variant of the pseudo-Siamese network, one of the deep learning algorithms, that does not need retraining to detect and classify untrained class targets. Our algorithm detects and classifies targets with only one shot. In addition, our algorithm does not need preprocessing. We verify our algorithm with Raman spectra measured using a Raman spectrometer.

Recently, geometrical effects in micro-/nanomaterials have gained great attention of researchers in diverse scientific and engineering fields because unique physical properties can be achieved not only by the micro-/nanosize effects of materials but also by adjustment of the geometrical parameters. In this work, we investigated heat-transfer in a perfectly aligned microwire array structure. When the structure has a very narrow gap (g) between the wires, which is less than twice the mean free path of an ambient medium (g < $2{\lambda }_{\mathit{medium}}$ ), there is a significant suppression of conduction heat-loss due to the reduction of the “molecular diffusion,” resulting in the improved power-efficiency in joule-heating characteristics. To clearly understand the suppressed conduction heat-loss with regard to the gap reduction, we performed analytical calculation and finite-element-method (FEM) simulation. To experimentally investigate the phenomenon, we develop a setup that is a combination of a microstructure having various gaps and an air-pressure controllable chamber. Through the experiment, we verified that a remarkable improvement of joule-heating operation can be achieved with the structure having very small gaps (g < $2{\lambda }_{\mathit{medium}}$ ) due to the suppression of conduction heat-loss.

In this article, we devise a variant of the extended Kalman filter that can be generally applied to systems on manifolds with simplicity and low computational cost. We extend a given system on a manifold to an ambient open set in Euclidean space and modify the system such that the extended system is transversely stable on the manifold. Then, we apply the standard extended Kalman filter derived in Euclidean space to the modified dynamics. This method is efficient in terms of computation and accurate in comparison with the standard extended Kalman filter. It has the merit that we can apply various Kalman filters derived in Euclidean space including extended Kalman filters for state estimation for systems defined on manifolds. The proposed method is successfully applied to the rigid body attitude dynamics whose configuration space is the special orthogonal group in three dimensions.