arXiv subjects

Yu Yang

Publications and source records attributed to Yu Yang.

At least 253 records · Page 14Linked to original sources

arXiv2022

Multiparameter simultaneous optimal estimation with an SU(2) coding unitary evolution

In a ubiquitous $SU(2)$ dynamics, achieving the simultaneous optimal estimation of multiple parameters is significant but difficult. Using quantum control to optimize this $SU(2)$ coding unitary evolution is one of solutions. We propose a method, characterized by the nested cross-products of the coefficient vector $\mathbf{X}$ of $SU(2)$ generators and its partial derivative $\partial_\ell \mathbf{X}$, to investigate the control-enhanced quantum multiparameter estimation. Our work reveals that quantum control is not always functional in improving the estimation precision, which depends on the characterization of an $SU(2)$ dynamics with respect to the objective parameter. This characterization is quantified by the angle $α_\ell$ between $\mathbf{X}$ and $\partial_\ell \mathbf{X}$. For an $SU(2)$ dynamics featured by $α_\ell=π/2$, the promotion of the estimation precision can get the most benefits from the controls. When $α_\ell$ gradually closes to $0$ or $π$, the precision promotion contributed to by quantum control correspondingly becomes inconspicuous. Until a dynamics with $α_\ell=0$ or $π$, quantum control completely loses its advantage. In addition, we find a set of conditions restricting the simultaneous optimal estimation of all the parameters, but fortunately, which can be removed by using a maximally entangled two-qubit state as the probe state and adding an ancillary channel into the configuration. Lastly, a spin-$1/2$ system is taken as an example to verify the above-mentioned conclusions. Our proposal sufficiently exhibits the hallmark of control-enhancement in fulfilling the multiparameter estimation mission, and it is applicable to an arbitrary $SU(2)$ parametrization process.

quant-ph↗

arXiv2022

A Dimension-Insensitive Algorithm for Stochastic Zeroth-Order Optimization

This paper concerns a convex, stochastic zeroth-order optimization (S-ZOO) problem. The objective is to minimize the expectation of a cost function whose gradient is not directly accessible. For this problem, traditional optimization algorithms mostly yield query complexities that grow polynomially with dimensionality (the number of decision variables). Consequently, these methods may not perform well in solving massive-dimensional problems arising in many modern applications. Although more recent methods can be provably dimension-insensitive, almost all of them require arguably more stringent conditions such as everywhere sparse or compressible gradient. In this paper, we propose a sparsity-inducing stochastic gradient-free (SI-SGF) algorithm, which provably yields a dimension-free (up to a logarithmic term) query complexity in both convex and strongly convex cases. Such insensitivity to the dimensionality growth is proven, for the first time, to be achievable when neither gradient sparsity nor gradient compressibility is satisfied. Our numerical results demonstrate a consistency between our theoretical prediction and the empirical performance.

math.OC↗

arXiv2022

Which Style Makes Me Attractive? Interpretable Control Discovery and Counterfactual Explanation on StyleGAN

The semantically disentangled latent subspace in GAN provides rich interpretable controls in image generation. This paper includes two contributions on semantic latent subspace analysis in the scenario of face generation using StyleGAN2. First, we propose a novel approach to disentangle latent subspace semantics by exploiting existing face analysis models, e.g., face parsers and face landmark detectors. These models provide the flexibility to construct various criterions with very concrete and interpretable semantic meanings (e.g., change face shape or change skin color) to restrict latent subspace disentanglement. Rich latent space controls unknown previously can be discovered using the constructed criterions. Second, we propose a new perspective to explain the behavior of a CNN classifier by generating counterfactuals in the interpretable latent subspaces we discovered. This explanation helps reveal whether the classifier learns semantics as intended. Experiments on various disentanglement criterions demonstrate the effectiveness of our approach. We believe this approach contributes to both areas of image manipulation and counterfactual explainability of CNNs. The code is available at \url{https://github.com/prclibo/ice}.

cs.CV↗

arXiv2021

Value Activation for Bias Alleviation: Generalized-activated Deep Double Deterministic Policy Gradients

It is vital to accurately estimate the value function in Deep Reinforcement Learning (DRL) such that the agent could execute proper actions instead of suboptimal ones. However, existing actor-critic methods suffer more or less from underestimation bias or overestimation bias, which negatively affect their performance. In this paper, we reveal a simple but effective principle: proper value correction benefits bias alleviation, where we propose the generalized-activated weighting operator that uses any non-decreasing function, namely activation function, as weights for better value estimation. Particularly, we integrate the generalized-activated weighting operator into value estimation and introduce a novel algorithm, Generalized-activated Deep Double Deterministic Policy Gradients (GD3). We theoretically show that GD3 is capable of alleviating the potential estimation bias. We interestingly find that simple activation functions lead to satisfying performance with no additional tricks, and could contribute to faster convergence. Experimental results on numerous challenging continuous control tasks show that GD3 with task-specific activation outperforms the common baseline methods. We also uncover a fact that fine-tuning the polynomial activation function achieves superior results on most of the tasks.

cs.LG↗

arXiv2021

Comprehensive Performance Evaluation of LID Practices for the Sponge City Construction: A Case Study in Guangxi, China

Sponge city construction is a new concept of urban stormwater management, which can effectively relieve urban flooding, reduce non-point source pollution, and promote the usage of rainwater resources, often including the application of Low Impact Development (LID) techniques. Although 30 cities in China have been chosen to implement sponge city construction, there is a lack of a quantitative evaluation method to evaluate the environmental, economic, and social benefits of LID practices. This paper develops a comprehensive evaluation system to quantify the benefits of different combinations of LID units using the Storm Water Management Model (SWMM) and the Analytical Hierarchy Process (AHP) method. The performance of five LID design scenarios with different locations and sizes of the bio-retention facility, the grassed swale, the sunken green space, the permeable pavement, and the storage tank were analyzed for a sports center project in Guangxi, China. Results indicated that the green scenario that contains 34.5% of bio-retention facilities and 46.0% of sunken green spaces had the best comprehensive performance regarding meeting the requirements of 75% annual total runoff reduction and the attainment of good operation performance, rainwater utilization, landscape promotion, and ecological service functions, mainly because they are micro-scale and decentralized facilities that can manage stormwater at the source through the natural process. The optimal scenario was adopted to construct the project, and the proposed evaluation system can also be applied to optimal selection and performance effect evaluation of LID practices in other sponge city projects.

stat.AP↗

arXiv2021

Learning Foreground-Background Segmentation from Improved Layered GANs

Deep learning approaches heavily rely on high-quality human supervision which is nonetheless expensive, time-consuming, and error-prone, especially for image segmentation task. In this paper, we propose a method to automatically synthesize paired photo-realistic images and segmentation masks for the use of training a foreground-background segmentation network. In particular, we learn a generative adversarial network that decomposes an image into foreground and background layers, and avoid trivial decompositions by maximizing mutual information between generated images and latent variables. The improved layered GANs can synthesize higher quality datasets from which segmentation networks of higher performance can be learned. Moreover, the segmentation networks are employed to stabilize the training of layered GANs in return, which are further alternately trained with Layered GANs. Experiments on a variety of single-object datasets show that our method achieves competitive generation quality and segmentation performance compared to related methods.

cs.CV↗

arXiv2021

Subtrees and BC-subtrees of maximum degree no more than k in trees

The subtrees and BC-subtrees (subtrees where any two leaves are at even distance apart) have been extensively studied in recent years. Such structures, under special constraints on degrees, have applications in many fields. Through an approach based on generating functions, we present recursive algorithms for enumerating various subtrees and BC-subtrees of maximum degree $\leq k$ in trees. The algorithms are illustrated through detailed examples. We also briefly discuss, in trees, the densities of subtrees (resp.~BC-subtrees) of maximum degree $\leq k(\geq 2)$ among all subtrees (resp.~BC-subtrees).

math.CO↗

arXiv2021

Parity measurement in the strong dispersive regime of circuit quantum acoustodynamics

Mechanical resonators are emerging as an important new platform for quantum science and technologies. A large number of proposals for using them to store, process, and transduce quantum information motivates the development of increasingly sophisticated techniques for controlling mechanical motion in the quantum regime. By interfacing mechanical resonators with superconducting circuits, circuit quantum acoustodynamics (cQAD) can make a variety of important tools available for manipulating and measuring motional quantum states. Here we demonstrate direct measurements of the phonon number distribution and parity of nonclassical mechanical states. We do this by operating our system in the strong dispersive regime, where a superconducting qubit can be used to spectroscopically resolve phonon Fock states. These measurements are some of the basic building blocks for constructing acoustic quantum memories and processors. Furthermore, our results open the door to performing even more complex quantum algorithms using mechanical systems, such as quantum error correction and multi-mode operations.

quant-ph↗

arXiv2021

Influence of database noises to machine learning for spatiotemporal chaos

A new strategy, namely the "clean numerical simulation" (CNS), was proposed (J. Computational Physics, 418:109629, 2020) to gain reliable/convergent simulations (with negligible numerical noises) of spatiotemporal chaotic systems in a long enough interval of time, which provide us benchmark solution for comparison. Here we illustrate that machine learning (ML) can always give good enough fitting predictions of a spatiotemporal chaos by using, separately, two quite different training sets: one is the "clean database" given by the CNS with negligible numerical noises, the other is the "polluted database" given by the traditional algorithms in single/double precision with considerably large numerical noises. However, even in statistics, the ML predictions based on the "polluted database" are quite different from those based on the "clean database". It illustrates that the database noises have huge influences on ML predictions of some spatiotemporal chaos, even in statistics. Thus, we must use a "clean" database for machine learning of some spatiotemporal chaos. This surprising result might open a new door and possibility to study machine learning.

physics.comp-ph↗

arXiv2021

Language Scaling for Universal Suggested Replies Model

We consider the problem of scaling automated suggested replies for Outlook email system to multiple languages. Faced with increased compute requirements and low resources for language expansion, we build a single universal model for improving the quality and reducing run-time costs of our production system. However, restricted data movement across regional centers prevents joint training across languages. To this end, we propose a multi-task continual learning framework, with auxiliary tasks and language adapters to learn universal language representation across regions. The experimental results show positive cross-lingual transfer across languages while reducing catastrophic forgetting across regions. Our online results on real user traffic show significant gains in CTR and characters saved, as well as 65% training cost reduction compared with per-language models. As a consequence, we have scaled the feature in multiple languages including low-resource markets.

cs.CL↗

arXiv2021

Greenplum: A Hybrid Database for Transactional and Analytical Workloads

Demand for enterprise data warehouse solutions to support real-time Online Transaction Processing (OLTP) queries as well as long-running Online Analytical Processing (OLAP) workloads is growing. Greenplum database is traditionally known as an OLAP data warehouse system with limited ability to process OLTP workloads. In this paper, we augment Greenplum into a hybrid system to serve both OLTP and OLAP workloads. The challenge we address here is to achieve this goal while maintaining the ACID properties with minimal performance overhead. In this effort, we identify the engineering and performance bottlenecks such as the under-performing restrictive locking and the two-phase commit protocol. Next we solve the resource contention issues between transactional and analytical queries. We propose a global deadlock detector to increase the concurrency of query processing. When transactions that update data are guaranteed to reside on exactly one segment we introduce one-phase commit to speed up query processing. Our resource group model introduces the capability to separate OLAP and OLTP workloads into more suitable query processing mode. Our experimental evaluation on the TPC-B and CH-benCHmark benchmarks demonstrates the effectiveness of our approach in boosting the OLTP performance without sacrificing the OLAP performance.

cs.DB↗

arXiv2021

Quantum state transfer between two photons with polarization and orbital angular momentum via quantum teleportation technology

Quantum teleportation is a useful quantum information technology to transmit quantum states between different degrees of freedom. We here report a quantum state transfer experiment in the linear optical system, transferring a single photon state in the polarization degree of freedom (DoF) to another photon in the orbital angular momentum (OAM) quantum state via a biphoton OAM entangled channel. Our experimental method is based on quantum teleportation technology. The differences between ours and the original teleportation scheme is that the transfer state is known in ours, and our method is for different particles with different DoFs while the original one is for different particles with same DoF. Besides, our present experiment is implemented with a high Bell-efficiency since each of the four hybrid-entangled Bell states can be discriminated. We use six states of poles of the Bloch sphere to test our experiment, and the fidelity of the quantum state transfer is $91.8\pm1.3\%$.

quant-ph↗

arXiv2021

Comprehensible Counterfactual Explanation on Kolmogorov-Smirnov Test

The Kolmogorov-Smirnov (KS) test is popularly used in many applications, such as anomaly detection, astronomy, database security and AI systems. One challenge remained untouched is how we can obtain an explanation on why a test set fails the KS test. In this paper, we tackle the problem of producing counterfactual explanations for test data failing the KS test. Concept-wise, we propose the notion of most comprehensible counterfactual explanations, which accommodates both the KS test data and the user domain knowledge in producing explanations. Computation-wise, we develop an efficient algorithm MOCHE (for MOst CompreHensible Explanation) that avoids enumerating and checking an exponential number of subsets of the test set failing the KS test. MOCHE not only guarantees to produce the most comprehensible counterfactual explanations, but also is orders of magnitudes faster than the baselines. Experiment-wise, we present a systematic empirical study on a series of benchmark real datasets to verify the effectiveness, efficiency and scalability of most comprehensible counterfactual explanations and MOCHE.

cs.LG↗

arXiv2021

Optimal Sharing and Fair Cost Allocation of Community Energy Storage

This paper studies an energy storage (ES) sharing model which is cooperatively invested by multiple buildings for harnessing on-site renewable utilization and grid price arbitrage. To maximize the economic benefits, we jointly consider the ES sizing, operation, and cost allocation via a coalition game formulation. Particularly, we study a fair ex-post cost allocation based on nucleolus which addresses fairness by minimizing the minimal dissatisfaction of all the players. To overcome the exponential computation burden caused by the implicit characteristic function, we employ a constraint generation technique to gradually approach the unique nucleolus by leveraging the sparse problem structure. We demonstrate both the fairness and computational efficiency of the method through case studies, which are not provided by the existing Shapley approach or proportional method. Particularly, only a small fraction of characteristic function (less than 1% for 20 buildings) is required to achieve the cost allocation versus the exponential information required by Shapley approach. Though there exists a minor increase of computation over the proportional method, the proposed method can ensure fairness while the latter fails in some cases. Further, we demonstrate both the building-wise and community-wise economic benefits are enhanced with the ES sharing model over the individual ES (IES) model. Accordingly, the overall value of ES is considerably improved (about 1.83 times).

cs.GT↗

arXiv2021

Maximums of generalized Hasse-Witt invariants and their applications to anabelian geometry

Let $(X, D_{X})$ be an arbitrary pointed stable curve of topological type $(g_{X}, n_{X})$ over an algebraically closed field of characteristic $p>0$. We prove that the generalized Hasse-Witt invariants of prime-to-$p$ cyclic admissible coverings of $(X, D_{X})$ attain maximum. As applications, we obtain an anabelian formula for $(g_{X}, n_{X})$, and prove that the field structures associated to inertia subgroups of marked points can be reconstructed group-theoretically from open continuous homomorphisms of admissible fundamental groups. Moreover, the formula for maximum generalized Hasse-Witt invariants and the result concerning reconstructions of field structures play important roles in the theory of moduli spaces of fundamental groups developed by the author of the present paper.

math.AG↗

arXiv2021

Stochastic Optimal Control of HVAC system for Energy-efficient Buildings

The heating, ventilation and air-conditioning (HVAC) system accounts for substantial energy use in buildings, whereas a large group of occupants are still not actually feeling comfortable staying inside. This poses the issue of developing energy-efficient HVAC control, i.e., reduce energy use (cost) while simultaneously enhancing human comfort. This paper pursues the objective and studies the stochastic optimal HVAC control subject to uncertain thermal demand (i.e., the weather and occupancy etc). Particularly, we involve the elaborate predicted mean vote (PMV) thermal comfort model in the optimization. The problem is computationally challenging due to the non-linear and non-analytical constraints imposed by the system dynamics and PMV model. We make the following contributions to address it. First, we formulate the problem as a Markov decision process (MDP) which is a desirable modeling technique capable of handling the complexities. Second, we propose a gradient-based learning (GB-L) method for progressively learning a stochastic control policy off-line and store it for on-line execution. Third, we prove the learning method converge to the optimal policies theoretically, and its performance (i.e., energy cost, thermal comfort and on-line computation) for HVAC control via simulations. The comparisons with the existing model predictive control based relaxation (MPC-R) method which is assumed with accurate future information and supposed to provide the near-optimal bounds, show that though there exists some performance loss in energy cost reduction (i.e., 6.5%), the proposed method can enable efficient on-line implementation (less than 1 second) and provide high probability of thermal comfort under uncertainties.

eess.SY↗

arXiv2021

Selling Renewable Utilization Service to Consumers via Cloud Energy Storage

This paper proposes a cloud energy storage (CES) model for enabling local renewable integration of building consumers (BCs). Different from most existing third-party based ES sharing models that the energy storage operator (ESO) gains profit by leasing energy or power capacity, our CES model allows the ESO to sell renewable utilization service (RUS) to its consumers, i.e., the total amount of local renewable generation shifted to supply their demand. Notably, we propose a quadratic price model for the ESO charging its consumers by their requested RUS and formulate their interactions as a Stackelberg game, which admits an equilibrium. We prove the CES model outperforms individual ES (IES) model in social welfare. Besides, we study the performance of the CES model compared with the IES model and an existing ES sharing model (referring to VES model) via case studies. We demonstrate the CES model can provide 2-4 times profit to the ESO than the VES model. Meanwhile, higher cost reduction for the BCs are secured by the CES model. Moreover, we show the CES model can achieve near social optima and high ES efficiency (i.e., utilization) which are not provided by the other ES models.

cs.GT↗

arXiv2021

Distributed Control of Multi-zone HVAC Systems Considering Indoor Air Quality

This paper studies a scalable control method for multi-zone heating, ventilation and air-conditioning (HVAC) systems to optimize the energy cost for maintaining thermal comfort and indoor air quality (IAQ) (represented by CO2) simultaneously. This problem is computationally challenging due to the complex system dynamics, various spatial and temporal couplings as well as multiple control variables to be coordinated. To address the challenges, we propose a two-level distributed method (TLDM) with a upper level and lower level control integrated. The upper level computes zone mass flow rates for maintaining zone thermal comfort with minimal energy cost, and then the lower level strategically regulates zone mass flow rates and the ventilation rate to achieve IAQ while preserving the near energy saving performance of upper level. As both the upper and lower level computation are deployed in a distributed manner, the proposed method is scalable and computationally efficient. The near-optimal performance of the method in energy cost saving is demonstrated through comparison with the centralized method. In addition, the comparisons with the existing distributed method show that our method can provide IAQ with only little increase of energy cost while the latter fails. Moreover, we demonstrate our method outperforms the demand controlled ventilation strategies (DCVs) for IAQ management with about 8-10% energy cost reduction.

eess.SY↗