Yingce Xia, Tao Qin, Nenghai Yu, Tie-Yan Liu
We study the existence of pure Nash equilibrium (PNE) for the mechanisms used in Internet services (e.g., online reviews and question-answer websites) to incentivize users to generate high-quality content. Most existing work assumes that users are homogeneous and have the same ability. However, real-world users are heterogeneous and their abilities can be very different from each other due to their diverse background, culture, and profession. In this work, we consider heterogeneous users with the following framework: (1) the users are heterogeneous and each of them has a private type indicating the best quality of the content she can generate; (2) there is a fixed amount of reward to allocate to the participated users. Under this framework, we study the existence of pure Nash equilibrium of several mechanisms composed by different allocation rules, action spaces, and information settings. We prove the existence of PNE for some mechanisms and the non-existence of PNE for some mechanisms. We also discuss how to find a PNE for those mechanisms with PNE either through a constructive way or a search algorithm.
Yingce Xia, Di He, Tao Qin, Liwei Wang, Nenghai Yu, Tie-Yan Liu, Wei-Ying Ma
While neural machine translation (NMT) is making good progress in the past two years, tens of millions of bilingual sentence pairs are needed for its training. However, human labeling is very costly. To tackle this training data bottleneck, we develop a dual-learning mechanism, which can enable an NMT system to automatically learn from unlabeled data through a dual-learning game. This mechanism is inspired by the following observation: any machine translation task has a dual task, e.g., English-to-French translation (primal) versus French-to-English translation (dual); the primal and dual tasks can form a closed loop, and generate informative feedback signals to train the translation models, even if without the involvement of a human labeler. In the dual-learning mechanism, we use one agent to represent the model for the primal task and the other agent to represent the model for the dual task, then ask them to teach each other through a reinforcement learning process. Based on the feedback signals generated during this process (e.g., the language-model likelihood of the output of a model, and the reconstruction error of the original sentence after the primal and dual translations), we can iteratively update the two models until convergence (e.g., using the policy gradient methods). We call the corresponding approach to neural machine translation \emph{dual-NMT}. Experiments show that dual-NMT works very well on English$\leftrightarrow$French translation; especially, by learning from monolingual data (with 10% bilingual data for warm start), it achieves a comparable accuracy to NMT trained from the full bilingual data for the French-to-English translation task.
Tao Qin, Jianhui Zhou, Junren Shi
We establish the general phonon dynamics of magnetic solids by incorporating the Mead-Truhlar correction in the Born-Oppenheimer approximation. The effective magnetic-field acting on the phonons naturally emerges, giving rise to the phonon Hall effect. A general formula of the intrinsic phonon Hall conductivity is obtained by using the corrected Kubo formula with the energy magnetization contribution incorporated properly. The resulting phonon Hall conductivity is fully determined by the phonon Berry curvature and the dispersions. Based on the formula, the topological phonon system could be rigorously defined. In the low temperature regime, we predict that the phonon Hall conductivity is proportional to $T^{3}$ for the ordinary phonon systems, while that for the topological phonon systems has the linear $T$ dependence with the quantized temperature coefficient.
Tao Qin, Walter Hofstetter
Time-periodically driven systems are a versatile toolbox for realizing interesting effective Hamiltonians. Heating, caused by excitations to high-energy states, is a challenge for experiments. While most setups address the relatively weakly-interacting regime so far, it is of general interest to study heating in strongly correlated systems. Using Floquet dynamical mean-field theory, we study non-equilibrium steady states (NESS) in the Falicov-Kimball model, with time-periodically driven kinetic energy or interaction. We systematically investigate the nonequilibrium properties of the NESS. For a driven kinetic energy, we show that resonant tunneling, where the interaction is an integer multiple of the driving frequency, plays an important role in the heating. In the strongly correlated regime, we show that this can be well understood using Fermi\textquoteright s golden rule and the Schrieffer-Wolff transformation for a time-periodically driven system. We furthermore demonstrate that resonant tunneling can be used to control the population of Floquet states to achieve "photo-doping". For driven interactions introduced by an oscillating magnetic field near a Feshbach resonance widely adopted, we find that the double occupancy is strongly modulated. Our calculations apply to shaken ultracold atom systems, and to solid state systems in a spatially uniform but time-dependent electric field. They are also closely related to lattice modulation spectroscopy. Our calculations are helpful to understand the latest experiments on strongly correlated Floquet systems.
Meisheng Zhao, Tao Qin, Yongde Zhang
We first derive for the general form of the fidelity for various bosonic channels. Thereby we give the fidelity of different quantum bosonic channel, possibly with product input and entangled input respectively, as examples. The properties of the fidelity are carefully examined.
Tao Qin
In type A, Kleshchev-Ram-Mathas realize Specht modules as quotient of Permutation modules, in this paper, we construct a Specht filtration of Permutation modules indexed by hook partition in affine type A; and construct a generalized Specht filtration of Permutation modules indexed by any partition in linear quiver case.
Tao Qin, Pengfei Zhang, Guoao Yang
Laser irradiation, as a versatile tool to tune topological properties of electronic systems, is under intensive studies. Experimentally, laser irradiation induced anomalous Hall effect in graphene has been observed (McIver et al., Nat. Phys. 16, 38 (2020)). Disorder is ubiquitous in real materials, and it has been shown that diagonal disorders, i.e., onsite disorder, can enhance topological properties of time-periodically driven quantum materials (Titum et al., Phys. Rev. Lett. 114, 056801 (2015)). Here, we investigate circularly polarized laser irradiated graphene with non-diagonal disorders, i.e., disordered tunneling, and find that disorder can induce nontrivial topological properties, characterized by Bott index and the real-space Chern number. Moreover, we show that one can turn on the laser irradiation non-adiabatically to drive the disordered graphene into non-trivial topological phase. It is a scheme which is especially interesting for experimental implementations.
Binyi Chen, Tao Qin, Tie-Yan Liu
We incorporate signaling scheme into Ad Auction setting, to achieve better welfare and revenue while protect users' privacy. We propose a new \emph{$K$-anonymous signaling scheme setting}, prove the hardness of the corresponding welfare/revenue maximization problem, and finally propose the algorithms to approximate the optimal revenue or welfare.
Tao Qin, Tie-Yan Liu
LETOR is a package of benchmark data sets for research on LEarning TO Rank, which contains standard features, relevance judgments, data partitioning, evaluation tools, and several baselines. Version 1.0 was released in April 2007. Version 2.0 was released in Dec. 2007. Version 3.0 was released in Dec. 2008. This version, 4.0, was released in July 2009. Very different from previous versions (V3.0 is an update based on V2.0 and V2.0 is an update based on V1.0), LETOR4.0 is a totally new release. It uses the Gov2 web page collection (~25M pages) and two query sets from Million Query track of TREC 2007 and TREC 2008. We call the two query sets MQ2007 and MQ2008 for short. There are about 1700 queries in MQ2007 with labeled documents and about 800 queries in MQ2008 with labeled documents. If you have any questions or suggestions about the datasets, please kindly email us (letor@microsoft.com). Our goal is to make the dataset reliable and useful for the community.
Tao Qin, Michele Fabrizio, S. Shahab Naghavi, Erio Tosatti
The damaging effect of strong electron-electron repulsion on regular, electron-phonon %$s$-wave superconductivity is a standard tenet. In spite of that, an increasing number of compounds such as fullerides and more recently alkali-doped aromatics exhibit %$s$-wave or presumably $s$ wave superconductivity despite very narrow bands and very strong electron repulsion. Here, we explore superconducting solutions of a model Hamiltonian inspired by the electronic structure of alkali doped aromatics. The model is a two-site, two-narrow-band metal with a single intersite phonon, leading to attraction-mediated, two-order parameter superconductivity. On top of that, the model includes a repulsive on-site Hubbard $U$, whose effect on the superconductivity we study. Starting within mean field, we find that $s \pm$ superconductivity is the best solution surviving the presence of $U$, whose effect is canceled out by the opposite signs of the two order parameters. The correlated Gutzwiller study that follows is necessary because without electron correlations the superconducting state would in this model be superseded by an antiferromagnetic insulating state with lower energy. The Gutzwiller correlations lower the energy of the metallic state, with the consequence that the $s \pm$ superconducting state is stabilized and even strengthened for small Hubbard $U$.
Weidong Ma, Bo Zheng, Tao Qin, Pingzhong Tang, Tie-Yan Liu
In this work, we study the problem of online mechanism design for resources allocation and pricing in cloud computing (RAPCC). We show that in general the allocation problems in RAPCC are NP-hard, and therefore we focus on designing dominant-strategy incentive compatible (DSIC) mechanisms with good competitive ratios compared to the offline optimal allocation (with the prior knowledge about the future jobs). We propose two kinds of DSIC online mechanisms. The first mechanism, which is based on a greedy allocation rule and leverages a priority function for allocation, is very fast and has a tight competitive bound. We discuss several priority functions including exponential and linear priority functions, and show that the former one has a better competitive ratio. The second mechanism, which is based on a dynamic program for allocation, also has a tight competitive ratio and performs better than the first one when the maximum demand of cloud customers is close to the capacity of the cloud provider.
Tao Qin, Meisheng Zhao, Yongde Zhang
Quantum communications using continuous variables are quite mature experimental techniques and the relevant theories have been extensively investigated with various methods. In this paper, we study the continuous variable quantum channels from a different angle, i.e., by exploring master equations. And we finally give explicitly the capacity of the channel we are studying. By the end of this paper, we derive the criterion for the optimal capacities of the Gaussian channel versus its fidelity.
Tao Qin
We develop a combinatorial framework for the subdivision map -- introduced by Maksimau, Mathas and Tubbenhauer -- between the KLR(W) algebras of type $A^{(1)}_{e-1}$ and type $A^{(1)}_{e}$, which provides a partial categorification of the runner removal theorems.
Tao Qin
We study the set $W_{r,e,w}\ $ of dominant weights of $\mathfrak{sl}_r$ arising from partitions of fixed $e$-weight $w$. For $e$-cores, we show that $W_{r,e,0}\ $ decomposes as a disjoint union of simplices indexed by compositions of $r$. For general $w$, we prove that $W_{r,e,w}\ $ is a disjoint union of copies of these simplices, with multiplicities determined by the corresponding quotient data, yielding in particular a closed counting formula for $|W_{r,e,w}\ |\ $. The geometry gives rise to the stingray patterns appearing in the title. More generally, it yields a natural labeling of the dominant $e$-alcoves meeting $W_{r,e,w}\ $ by weak compositions of $w$, together with a compatible partial action of the affine Weyl group via wall crossing. Finally, we give an explicit alcove-geometric proof of the empty runner removal theorem for Iwahori-Hecke algebras.
Junjie Li, Sotetsu Koyamada, Qiwei Ye, Guoqing Liu, Chao Wang, Ruihan Yang, Li Zhao, Tao Qin, Tie-Yan Liu, Hsiao-Wuen Hon
Artificial Intelligence (AI) has achieved great success in many domains, and game AI is widely regarded as its beachhead since the dawn of AI. In recent years, studies on game AI have gradually evolved from relatively simple environments (e.g., perfect-information games such as Go, chess, shogi or two-player imperfect-information games such as heads-up Texas hold'em) to more complex ones (e.g., multi-player imperfect-information games such as multi-player Texas hold'em and StartCraft II). Mahjong is a popular multi-player imperfect-information game worldwide but very challenging for AI research due to its complex playing/scoring rules and rich hidden information. We design an AI for Mahjong, named Suphx, based on deep reinforcement learning with some newly introduced techniques including global reward prediction, oracle guiding, and run-time policy adaptation. Suphx has demonstrated stronger performance than most top human players in terms of stable rank and is rated above 99.99% of all the officially ranked human players in the Tenhou platform. This is the first time that a computer program outperforms most top human players in Mahjong.
Weihao Kong, Jian Li, Tao Qin, Tie-Yan Liu
Group-buying websites represented by Groupon.com are very popular in electronic commerce and online shopping nowadays. They have multiple slots to provide deals with significant discounts to their visitors every day. The current user traffic allocation mostly relies on human decisions. We study the problem of automatically allocating the user traffic of a group-buying website to different deals to maximize the total revenue and refer to it as the Group-buying Allocation Problem (\GAP). The key challenge of \GAP\ is how to handle the tipping point (lower bound) and the purchase limit (upper bound) of each deal. We formulate \GAP\ as a knapsack-like problem with variable-sized items and majorization constraints. Our main results for \GAP\ can be summarized as follows. (1) We first show that for a special case of \GAP, in which the lower bound equals the upper bound for each deal, there is a simple dynamic programming-based algorithm that can find an optimal allocation in pseudo-polynomial time. (2) The general case of \GAP\ is much more difficult than the special case. To solve the problem, we first discover several structural properties of the optimal allocation, and then design a two-layer dynamic programming-based algorithm leveraging those properties. This algorithm can find an optimal allocation in pseudo-polynomial time. (3) We convert the two-layer dynamic programming based algorithm to a fully polynomial time approximation scheme (FPTAS), using the technique developed in \cite{ibarra1975fast}, combined with some careful modifications of the dynamic programs. Besides these results, we further investigate some natural generalizations of \GAP, and propose effective algorithms.
Tao Qin, Meisheng Zhao, Yongde Zhang
May 19, 2006·quant-ph·PDF The process of quantum teleportation can be considered as a quantum channel. The exact classical capacity of the continuous variable teleportation channel is given. Also, the channel fidelity is derived. Consequently, the properties of the continuous variable quantum teleportation are discussed and interesting results are obtained.
Qin Tao, Shuowen Zhang, Caijun Zhong, Rui Zhang
In this letter, we consider a multicast system where a single-antenna transmitter sends a common message to multiple single-antenna users, aided by an intelligent reflecting surface (IRS) equipped with $N$ passive reflecting elements. Prior works on IRS have mostly assumed the availability of channel state information (CSI) for designing its passive beamforming. However, the acquisition of CSI requires substantial training overhead that increases with $N$. In contrast, we propose in this letter a novel \emph{random passive beamforming} scheme, where the IRS performs independent random reflection for $Q\geq 1$ times in each channel coherence interval without the need of CSI acquisition. For the proposed scheme, we first derive a closed-form approximation of the outage probability, based on which the optimal $Q$ with best outage performance can be efficiently obtained. Then, for the purpose of comparison, we derive a lower bound of the outage probability with traditional CSI-based passive beamforming. Numerical results show that a small $Q$ is preferred in the high-outage regime (or with high rate target) and the optimal $Q$ becomes larger as the outage probability decreases (or as the rate target decreases). Moreover, the proposed scheme significantly outperforms the CSI-based passive beamforming scheme with training overhead taken into consideration when $N$ and/or the number of users are large, thus offering a promising CSI-free alternative to existing CSI-based schemes.
Yingce Xia, Tao Qin, Wei Chen, Jiang Bian, Nenghai Yu, Tie-Yan Liu
Many supervised learning tasks are emerged in dual forms, e.g., English-to-French translation vs. French-to-English translation, speech recognition vs. text to speech, and image classification vs. image generation. Two dual tasks have intrinsic connections with each other due to the probabilistic correlation between their models. This connection is, however, not effectively utilized today, since people usually train the models of two dual tasks separately and independently. In this work, we propose training the models of two dual tasks simultaneously, and explicitly exploiting the probabilistic correlation between them to regularize the training process. For ease of reference, we call the proposed approach \emph{dual supervised learning}. We demonstrate that dual supervised learning can improve the practical performances of both tasks, for various applications including machine translation, image processing, and sentiment analysis.
Wenkui Ding, Tao Wu, Tao Qin, Tie-Yan Liu
The Generalized Second Price auction (GSP) has been widely used by search engines to sell ad slots. Previous studies have shown that the pure Price Of Anarchy (POA) of GSP is 1.25 when there are two ad slots and 1.259 when three ad slots. For the cases with more than three ad slots, however, only some untight upper bounds of the pure POA were obtained. In this work, we improve previous results in two aspects: (1) We prove that the pure POA for GSP is 1.259 when there are four ad slots, and (2) We show that the pure POA for GSP with more than four ad slots is also 1.259 given the bidders are ranked according to a particular permutation.