Nonlinear topological photonic and phononic systems have recently aroused intense interests in exploring new phenomena that have no counterparts in electronic systems. The squeezed bosonic interaction in these systems is particularly interesting, bec
ause it can modify the vacuum fluctuations of topological states, drive them into instabilities, and lead to topological parametric lasers. However, these phenomena remain experimentally elusive because of limited nonlinearities in most existing topological bosonic systems. Here, we experimentally realized topological parametric lasers based on nonlinear nanoelectromechanical Dirac-vortex cavities with strong squeezed interaction. Specifically, we parametrically drove the Dirac-vortex cavities to provide phase-sensitive amplification for topological phonons, and observed phonon lasing above the threshold. Additionally, we confirmed that the lasing frequency is robust against fabrication disorders and that the free spectral range defies the universal inverse scaling law with increased cavity size, which benefit the realization of large-area single-mode lasers. Our results represent an important advance in experimental investigations of topological physics with large bosonic nonlinearities and parametric gain.
Answer selection, which is involved in many natural language processing applications such as dialog systems and question answering (QA), is an important yet challenging task in practice, since conventional methods typically suffer from the issues of
ignoring diverse real-world background knowledge. In this paper, we extensively investigate approaches to enhancing the answer selection model with external knowledge from knowledge graph (KG). First, we present a context-knowledge interaction learning framework, Knowledge-aware Neural Network (KNN), which learns the QA sentence representations by considering a tight interaction with the external knowledge from KG and the textual information. Then, we develop two kinds of knowledge-aware attention mechanism to summarize both the context-based and knowledge-based interactions between questions and answers. To handle the diversity and complexity of KG information, we further propose a Contextualized Knowledge-aware Attentive Neural Network (CKANN), which improves the knowledge representation learning with structure information via a customized Graph Convolutional Network (GCN) and comprehensively learns context-based and knowledge-based sentence representation via the multi-view knowledge-aware attention mechanism. We evaluate our method on four widely-used benchmark QA datasets, including WikiQA, TREC QA, InsuranceQA and Yahoo QA. Results verify the benefits of incorporating external knowledge from KG, and show the robust superiority and extensive applicability of our method.
Monolayer WTe2 is predicted to be a quantum spin Hall insulator (QSHI) and electron transport along its edges has been experimentally observed. However, the smoking gun of QSHI, spin momentum locking of the edge electrons, has not been experimentally
demonstrated. We propose a model to establish the relationship between the anisotropic magnetoresistance (AMR) and spin orientation of the helical electrons in WTe2. Based on the predictions of the model, angular dependent magnetoresistance measurements were carried out. The experimental results fully supported the model and the spin orientation of the helical edge electrons was determined. Our results not only demonstrate that WTe2 is indeed a QSHI, but also suggest a convenient method to determine the spin orientation of other QSHIs.
High-order interactive features capture the correlation between different columns and thus are promising to enhance various learning tasks on ubiquitous tabular data. To automate the generation of interactive features, existing works either explicitl
y traverse the feature space or implicitly express the interactions via intermediate activations of some designed models. These two kinds of methods show that there is essentially a trade-off between feature interpretability and search efficiency. To possess both of their merits, we propose a novel method named Feature Interaction Via Edge Search (FIVES), which formulates the task of interactive feature generation as searching for edges on the defined feature graph. Specifically, we first present our theoretical evidence that motivates us to search for useful interactive features with increasing order. Then we instantiate this search strategy by optimizing both a dedicated graph neural network (GNN) and the adjacency tensor associated with the defined feature graph. In this way, the proposed FIVES method simplifies the time-consuming traversal as a typical training course of GNN and enables explicit feature generation according to the learned adjacency tensor. Experimental results on both benchmark and real-world datasets show the advantages of FIVES over several state-of-the-art methods. Moreover, the interactive features identified by FIVES are deployed on the recommender system of Taobao, a worldwide leading e-commerce platform. Results of an online A/B testing further verify the effectiveness of the proposed method FIVES, and we further provide FIVES as AI utilities for the customers of Alibaba Cloud.
The textbook-accepted formulation of electromagnetic force was proposed by Lorentz in the 19th century, but its validity has been challenged due to incompatibility with the special relativity and momentum conservation. The Einstein-Laub formulation,
which can reconcile those conflicts, was suggested as an alternative to the Lorentz formulation. However, intense debates on the exact force are still going on due to lack of experimental evidence. Here, we report the first experimental investigation of angular symmetry of optical force inside a solid dielectric, aiming to distinguish the two formulations. The experiments surprisingly show that the optical force exerted by a Gaussian beam has components with the angular mode number of both 2 and 0, which cannot be explained solely by the Lorentz or the Einstein-Laub formulation. Instead, we found a modified Helmholtz theory by combining the Lorentz force with additional electrostrictive force could explain our experimental results. Our results represent a fundamental leap forward in determining the correct force formulation, and will update the working principles of many applications involving electromagnetic forces.
Discrete degrees of freedom, such as spin and orbital, can provide intriguing strategies to manipulate electrons, photons, and phonons. With a spin degree of freedom, topological insulators have stimulated intense interests in condensed-matter physic
s, optics, acoustics, and mechanics. However, orbital as another fundamental attribute in crystals has seldom been investigated in topological insulators. Here, we invent a new type of topological insulators with an auxiliary orbital degree of freedom on a nanomechanical platform. We experimentally realized nanomechanical topological insulators where the orbital can arbitrarily be manipulated by the crystal. Harnessing this unique feature, we demonstrated adiabatic transition between distinct topological edge states, which is a crucial functionality for complicated systems that involve distinct topological edge channels. Beyond the one-dimensional edge states, we further constructed zero-dimensional Dirac-vortex states. Our results have unveiled unprecedented strategies to manipulate topological phase transitions and to study topological phases of matter on an integrated platform.
The recent realizations of topological valley phase in photonic crystal, an analog of gapped valleytronic materials in electronic system, are limited to the valley Chern number of one. In this letter, we present a new type of valley phase that can ha
ve large valley Chern number of two or three. The valley phase transitions between the different valley Chern numbers (from one to three) are realized by changing the configuration of the unit cell. We demonstrate that these new topological phases can guide the wave propagation robustly along the domain wall of sharp bent. Our results are promising for the exploration of new topological phenomena in photonic systems.
Biomedical researchers usually study the effects of certain exposures on disease risks among a well-defined population. To achieve this goal, the gold standard is to design a trial with an appropriate sample from that population. Due to the high cost
of such trials, usually the sample size collected is limited and is not enough to accurately estimate some exposures effect. In this paper, we discuss how to leverage the information from external `big data (data with much larger sample size) to improve the estimation accuracy at the risk of introducing small bias. We proposed a family of weighted estimators to balance the bias increase and variance reduction when including the big data. We connect our proposed estimator to the established penalized regression estimators. We derive the optimal weights using both second order and higher order asymptotic expansions. Using extensive simulation studies, we showed that the improvement in terms of mean square error (MSE) for the regression coefficient can be substantial even with finite sample sizes and our weighted method outperformed the existing methods such as penalized regression and James Steins approach. Also we provide theoretical guarantee that the proposed estimators will never lead to asymptotic MSE larger than the maximum likelihood estimator using small data only in general. We applied our proposed methods to the Asia Cohort Consortium China cohort data to estimate the relationships between age, BMI, smoking, alcohol use and mortality.
The current paper deals with limited-budget output consensus for descriptor multiagent systems with two types of switching communication topologies; that is, switching connected ones and jointly connected ones. Firstly, a singular dynamic output feed
back control protocol with switching communication topologies is proposed on the basis of the observable decomposition, where an energy constraint is involved and protocol states of neighboring agents are utilized to derive a new two-step design approach of gain matrices. Then, limited-budget output consensus problems are transformed into asymptotic stability ones and a valid candidate of the output consensus function is determined. Furthermore, sufficient conditions for limited-budget output consensus design for two types of switching communication topologies are proposed, respectively. Finally, two numerical simulations are shown to demonstrate theoretical conclusions.
Valley pseudospin, a new degree of freedom in photonic lattices, provides an intriguing way to manipulate photons and enhance the robustness of optical networks. Here we experimentally demonstrated topological waveguiding, refracting, resonating, and
routing of valley-polarized photons in integrated circuits. Specifically, we show that at the domain wall between photonic crystals of different topological valley phases, there exists a topologically protected valley kink state that is backscattering-free at sharp bends and terminals. We further harnessed these valley kink states for constructing high-Q topological photonic crystal cavities with tortuously shaped cavity geometries. We also demonstrated a novel optical routing scheme at an intersection of multiple valley kink states, where light splits counterintuitively due to the valley pseudospin of photons. These results will not only lead to robust optical communication and signal processing, but also open the door for fundamental research of topological photonics in areas such as lasing, quantum photon-pair generation, and optomechanics.
