Virtual Classification: Modulating Domain-Specific Knowledge for Multidomain Crowd Counting.

Multidomain crowd counting aims to learn a general model for multiple diverse datasets. However, deep networks prefer modeling distributions of the dominant domains instead of all domains, which is known as domain bias. In this study, we propose a simple-yet-effective modulating domain-specific knowledge network (MDKNet) to handle the domain bias issue in multidomain crowd counting. MDKNet is achieved by employing the idea of "modulating", enabling deep network balancing and modeling different distributions of diverse datasets with little bias. Specifically, we propose an instance-specific batch normalization (IsBN) module, which serves as a base modulator to refine the information flow to be adaptive to domain distributions. To precisely modulating the domain-specific information, the domain-guided virtual classifier (DVC) is then introduced to learn a domain-separable latent space. This space is employed as an input guidance for the IsBN modulator, such that the mixture distributions of multiple datasets can be well treated. Extensive experiments performed on popular benchmarks, including Shanghai-tech A/B, QNRF, and NWPU validate the superiority of MDKNet in tackling multidomain crowd counting and the effectiveness for multidomain learning. Code is available at https://github.com/csguomy/MDKNet.

Landauer-QFLPS Model for Mixed Schottky-Ohmic Contact Two-Dimensional Transistors.

Two-dimensional material-based field-effect transistors (2DM-FETs) are playing a revolutionary role in electronic devices. However, before electronic design automation (EDA) for 2DM-FETs can be achieved, it remains necessary to determine how to incorporate contact transports into model. Reported methods compromise between physical intelligibility and model compactness due to the heterojunction nature. To address this, quasi-Fermi-level phase space theory (QFLPS) is generalized to incorporate contact transports using the Landauer formula. It turns out that the Landauer-QFLPS model effectively overcomes the issue of concern. The proposed new formula can describe 2DM-FETs with Schottky or Ohmic contacts with superior accuracy and efficiency over previous methods, especially when describing non-monotonic drain conductance characteristics. A three-bit threshold inverter quantizer (TIQ) circuit is fabricated using ambipolar black phosphorus and it is demonstrated that the model accurately predicts circuit performance. The model could be very effective and valuable in the development of 2DM-FET-based integrated circuits.

DFT-1/2 and shell DFT-1/2 methods: electronic structure calculation for semiconductors at LDA complexity.

It is known that the Kohn-Sham eigenvalues do not characterize experimental excitation energies directly, and the band gap of a semiconductor is typically underestimated by local density approximation (LDA) of density functional theory (DFT). An embarrassing situation is that one usually uses LDA+Ufor strongly correlated materials with rectified band gaps, but for non-strongly-correlated semiconductors one has to resort to expensive methods like hybrid functionals orGW. In spite of the state-of-the-art meta-generalized gradient approximation functionals like TB-mBJ and SCAN, methods with LDA-level complexity to rectify the semiconductor band gaps are in high demand. DFT-1/2 stands as a feasible approach and has been more widely used in recent years. In this work we give a detailed derivation of the Slater half occupation technique, and review the assumptions made by DFT-1/2 in semiconductor band structure calculations. In particular, the self-energy potential approach is verified through mathematical derivations. The aims, features and principles of shell DFT-1/2 for covalent semiconductors are also accounted for in great detail. Other developments of DFT-1/2 including conduction band correction, DFT+A-1/2, empirical formula for the self-energy potential cutoff radius, etc, are further reviewed. The relations of DFT-1/2 to hybrid functional, sX-LDA,GW, self-interaction correction, scissor's operator as well as DFT+Uare explained. Applications, issues and limitations of DFT-1/2 are comprehensively included in this review.

Two-stage amplification of an ultrasensitive MXene-based intelligent artificial eardrum.

We report an artificial eardrum using an acoustic sensor based on two-dimensional MXene (Ti3C2Tx), which mimics the function of a human eardrum for realizing voice detection and recognition. Using MXene with a large interlayer distance and micropyramid polydimethylsiloxane arrays can enable a two-stage amplification of pressure and acoustic sensing. The MXene artificial eardrum shows an extremely high sensitivity of 62 kPa-1 and a very low detection limit of 0.1 Pa. Notably, benefiting from the ultrasensitive MXene eardrum, the machine-learning algorithm for real-time voice classification can be realized with high accuracy. The 280 voice signals are successfully classified for seven categories, and a high accuracy of 96.4 and 95% can be achieved by the training dataset and the test dataset, respectively. The current results indicate that the MXene artificial intelligent eardrum shows great potential for applications in wearable acoustical health care devices.

Gate-Tunable Negative Differential Resistance Behaviors in a hBN-Encapsulated BP-MoS2 Heterojunction.

Two-dimensional (2D) heterostructures show great potential in achieving negative differential resistance (NDR) effects by Esaki diodes and or resonant tunneling diodes. However, most of the reported Esaki diode-based NDR devices realized by bulk 2D films lack sufficient gate tunability, and the tuning of NDR behavior from appearing to vanishing remains elusive. Here, a gate-tunable NDR device is reported based on a vertically stacked black phosphorus (BP) and molybdenum disulfide (MoS2) thin 2D heterojunction. At room temperature, a rectifying ratio of ∼6 orders of magnitude from a reverse rectifying diode to a forward rectifying diode by gate modulation is obtained. Through analyzing the temperature-dependent electrical properties, the tunneling mechanism at a certain gate voltage range is revealed. Moreover, the switchable and continuously gate-tunable NDR behavior is realized with a maximum peak-to-valley ratio of 1.23 at 77 K, as shown in the IDS mappings by sweeping VDS under different VGS. In addition, a compact model for gate-tunable NDR behavior in 2D heterostructures is proposed. To our best knowledge, this is the first demonstration of NDR behavior in BP-MoS2 heterostructures. Consequently, this work sheds light on the gate-tunable NDR devices and reconfigurable logic devices for realizing ternary and reconfigurable logic systems.

Post-Stroke Microglia Induce Sirtuin2 Expression to Suppress the Anti-inflammatory Function of Infiltrating Regulatory T Cells.

Ischemic stroke is among the leading causes of death and disability across the globe. Post-stroke neuroinflammation contributes to the pathophysiology of ischemic stroke in the acute phase through damaging neurons in the penumbra region. Infiltrating regulatory T cells (Treg cells) provide neuronal protection in ischemic brains. In the current study using a mouse-transient middle cerebral artery occlusion (MCAO) model, we characterized the changes of sirtuin expression in infiltrating Treg cells in the acute phase of ischemia. We found that Sirt2 was remarkably upregulated in infiltrating Treg cells at day 3 post-MCAO. In vitro inhibition of Sirt2 activity enhanced the expression of immunosuppression-associated molecules including forkhead box P3 (Foxp3) in Treg cells. Using a lentiviral system to express exogenous Sirt2 in Treg cells, we found that Sirt2 weakened the anti-inflammatory effect of Treg cells on pro-inflammatory macrophages. Additionally, post-MCAO microglia increased Sirt2 expression in Treg cells in a cell-to-cell contact manner. We further found that microglia remarkably induced hypoxia-inducible factor 1-alpha (HIF-1α) expression in Treg cells, and inhibition of HIF-1α abolished microglia-induced Sirt2 upregulation. Collectively, we discovered a novel mechanism by which the immunoregulatory activity of infiltrating Treg cells is modulated after ischemia.

Two-Mode MoS2 Filament Transistor with Extremely Low Subthreshold Swing and Record High On/Off Ratio.

With rapid development of integrated circuits, urgent requirements for a transistor with lower subthreshold swing (SS) and better contact properties are needed. To optimize the SS and contact issues, we propose a concept of molybdenum disulfide (MoS2) filament transistor with two modes. We successfully fabricated the proposed devices in a wafer-scale. Mode I can enable the device with extremely low SS down to 2.26 mV/dec by switching the contact filament between on and off while mode II can realize a record high on/off ratio of 2.6 × 109 by using filament as quasi-zero dimensional (quasi-0D) contact. Compared to conventional three-dimensional (3D) contact, quasi-0D contact using conductive filament improves the current density nearly 50 times. We also built a spice model to simulate the electrical behaviors, and the simulation results show an extremely low SS in mode I (using abrupt filament formation/rupture) and excellent quasi-0D contact in mode II. The two-mode MoS2 filament transistor can significantly improve the SS and contact comparing to those of the state-of-the-art transistors, which has the great potential to boost the development of the next generation mainstream transistors.