Search

In recommendation systems, the relevance and novelty of the final results are selected through a cascade system of Matching -> Ranking -> Strategy. The matching model serves as the starting point of the pipeline and determines the upper bound of the subsequent stages. Balancing the relevance and novelty of matching results is a crucial step in the design and optimization of recommendation systems, contributing significantly to improving recommendation quality. However, the typical matching algorithms have not simultaneously addressed the relevance and novelty perfectly. One main reason is that deep matching algorithms exhibit significant uncertainty when estimating items in the long tail (e.g., due to insufficient training samples) items.The uncertainty not only affects the training of the models but also influences the confidence in the index construction and beam search retrieval process of these models. This paper proposes the UICR (Uncertainty-based explore for Index Construction and Retrieval) algorithm, which introduces the concept of uncertainty modeling in the matching stage and achieves multi-task modeling of model uncertainty and index uncertainty. The final matching results are obtained by combining the relevance score and uncertainty score infered by the model. Experimental results demonstrate that the UICR improves novelty without sacrificing relevance on realworld industrial productive environments and multiple open-source datasets. Remarkably, online A/B test results of display advertising in Shopee demonstrates the effectiveness of the proposed algorithm., Comment: accepted by cikm2024

We extract key information of dark energy from current observations of BAO, OHD and $H_0$, and find hints of dynamical behaviour of dark energy. In particular, a dynamical dark energy model whose equation of state crosses $-1$ is favoured by observations. We also find that the Universe has started accelerating at a lower redshift than expected., Comment: 10 pages, 4 figures; version accepted to RAA

We develop methods to extract key dark energy information from cosmic distance measurements including the BAO scales and supernovae luminosity distances. Demonstrated using simulated datasets of the complete DESI, LSST and Roman surveys designed for BAO and SNe distance measurements, we show that using our method, the dynamical behaviour of the energy, pressure, equation of state (with its time derivative) of dark energy and the cosmic deceleration function can all be accurately recovered from high-quality data, which allows for robust diagnostic tests for dark energy models., Comment: 15 pages, 7 figures; version accepted to RAA

This paper introduces a new stochastic hybrid system (SHS) framework for contingency detection in modern power systems (MPS). The framework uses stochastic hybrid system representations in state space models to expand and facilitate capability of contingency detection. In typical microgrids (MGs), buses may contain various synchronous generators, renewable generators, controllable loads, battery systems, regular loads, etc. For development of SHS models in power systems, this paper introduces the concept of dynamic and non-dynamic buses. By converting a physical power grid into a virtual linearized state space model and representing contingencies as random switching of system structures and parameters, this paper formulates the contingency detection problem as a joint estimation problem of discrete event and continuous states in stochastic hybrid systems. This method offers unique advantages, including using common measurement signals on voltage and current synchrophasors to detect different types and locations of contingencies, avoiding expensive local direct fault measurements and detecting certain contingencies that cannot be directly measured. The method employs a small and suitably-designed probing signal to sustain the ability of persistent contingency detection. Joint estimation algorithms are presented with their proven convergence and reliability properties. Examples that use an IEEE 5-bus system demonstrate the main ideas and derivation steps. Simulation case studies on an IEEE 33-bus system are used for detecting transmission line faults and sensor interruptions., Comment: 12 pages, 10 figures. arXiv admin note: text overlap with arXiv:2401.16568

This paper introduces a stochastic hybrid system (SHS) framework in state space model to capture sensor, communication, and system contingencies in modern power systems (MPS). Within this new framework, the paper concentrates on the development of state estimation methods and algorithms to provide reliable state estimation under randomly intermittent and noisy sensor data. MPSs employ diversified measurement devices for monitoring system operations that are subject to random measurement errors and rely on communication networks to transmit data whose channels encounter random packet loss and interruptions. The contingency and noise form two distinct and interacting stochastic processes that have a significant impact on state estimation accuracy and reliability. This paper formulates stochastic hybrid system models for MPSs, introduces coordinated observer design algorithms for state estimation, and establishes their convergence and reliability properties. A further study reveals a fundamental design tradeoff between convergence rates and steady-state error variances. Simulation studies on the IEEE 5-bus system and IEEE 33-bus system are used to illustrate the modeling methods, observer design algorithms, convergence properties, performance evaluations, and impact sensor system selections., Comment: 15 pages, 9 figures

In the e-commerce advertising scenario, estimating the true probabilities (known as a calibrated estimate) on Click-Through Rate (CTR) and Conversion Rate (CVR) is critical. Previous research has introduced numerous solutions for addressing the calibration problem. These methods typically involve the training of calibrators using a validation set and subsequently applying these calibrators to correct the original estimated values during online inference. However, what sets e-commerce advertising scenarios apart is the challenge of multi-field calibration. Multi-field calibration requires achieving calibration in each field. In order to achieve multi-field calibration, it is necessary to have a strong data utilization ability. Because the quantity of pCTR specified range for a single field-value (such as user ID and item ID) sample is relatively small, this makes the calibrator more difficult to train. However, existing methods have difficulty effectively addressing these issues. To solve these problems, we propose a new method named Deep Ensemble Shape Calibration (DESC). In terms of business understanding and interpretability, we decompose multi-field calibration into value calibration and shape calibration. We introduce innovative basis calibration functions, which enhance both function expression capabilities and data utilization by combining these basis calibration functions. A significant advancement lies in the development of an allocator capable of allocating the most suitable calibrators to different estimation error distributions within diverse fields and values. We achieve significant improvements in both public and industrial datasets. In online experiments, we observe a +2.5% increase in CVR and +4.0% in GMV (Gross Merchandise Volume). Our code is now available at: https://github.com/HaoYang0123/DESC.

In online advertising scenario, sellers often create multiple creatives to provide comprehensive demonstrations, making it essential to present the most appealing design to maximize the Click-Through Rate (CTR). However, sellers generally struggle to consider users preferences for creative design, leading to the relatively lower aesthetics and quantities compared to Artificial Intelligence (AI)-based approaches. Traditional AI-based approaches still face the same problem of not considering user information while having limited aesthetic knowledge from designers. In fact that fusing the user information, the generated creatives can be more attractive because different users may have different preferences. To optimize the results, the generated creatives in traditional methods are then ranked by another module named creative ranking model. The ranking model can predict the CTR score for each creative considering user features. However, the two above stages are regarded as two different tasks and are optimized separately. In this paper, we proposed a new automated Creative Generation pipeline for Click-Through Rate (CG4CTR) with the goal of improving CTR during the creative generation stage. Our contributions have 4 parts: 1) The inpainting mode in stable diffusion is firstly applied to creative generation task in online advertising scene. A self-cyclic generation pipeline is proposed to ensure the convergence of training. 2) Prompt model is designed to generate individualized creatives for different user groups, which can further improve the diversity and quality. 3) Reward model comprehensively considers the multimodal features of image and text to improve the effectiveness of creative ranking task, and it is also critical in self-cyclic pipeline. 4) The significant benefits obtained in online and offline experiments verify the significance of our proposed method.

Software-defined satellite-terrestrial integrated networks (SDSTNs) are seen as a promising paradigm for achieving high resource flexibility and global communication coverage. However, low latency service provisioning is still challenging due to the fast variation of network topology and limited onboard resource at low earth orbit satellites. To address this issue, we study service provisioning in SDSTNs via joint optimization of virtual network function (VNF) placement and routing planning with network dynamics characterized by a time-evolving graph. Aiming at minimizing average service latency, the corresponding problem is formulated as an integer nonlinear programming under resource, VNF deployment, and time-slotted flow constraints. Since exhaustive search is intractable, we transform the primary problem into an integer linear programming by involving auxiliary variables and then propose a Benders decomposition based branch-and-cut (BDBC) algorithm. Towards practical use, a time expansion-based decoupled greedy (TEDG) algorithm is further designed with rigorous complexity analysis. Extensive experiments demonstrate the optimality of BDBC algorithm and the low complexity of TEDG algorithm. Meanwhile, it is indicated that they can improve the number of completed services within a configuration period by up to 58% and reduce the average service latency by up to 17% compared to baseline schemes., Comment: Accepted by IEEE Transactions on Wireless Communications

Adversarial poisoning attacks pose huge threats to various machine learning applications. Especially, the recent accumulative poisoning attacks show that it is possible to achieve irreparable harm on models via a sequence of imperceptible attacks followed by a trigger batch. Due to the limited data-level discrepancy in real-time data streaming, current defensive methods are indiscriminate in handling the poison and clean samples. In this paper, we dive into the perspective of model dynamics and propose a novel information measure, namely, Memorization Discrepancy, to explore the defense via the model-level information. By implicitly transferring the changes in the data manipulation to that in the model outputs, Memorization Discrepancy can discover the imperceptible poison samples based on their distinct dynamics from the clean samples. We thoroughly explore its properties and propose Discrepancy-aware Sample Correction (DSC) to defend against accumulative poisoning attacks. Extensive experiments comprehensively characterized Memorization Discrepancy and verified its effectiveness. The code is publicly available at: https://github.com/tmlr-group/Memorization-Discrepancy., Comment: accepted by ICML 2023

Adverse conditions typically suffer from stochastic hybrid weather degradations (e.g., rainy and hazy night), while existing image restoration algorithms envisage that weather degradations occur independently, thus may fail to handle real-world complicated scenarios. Besides, supervised training is not feasible due to the lack of a comprehensive paired dataset to characterize hybrid conditions. To this end, we have advanced the aforementioned limitations with two tactics: framework and data. First, we present a novel unified framework, dubbed RAHC, to Restore Arbitrary Hybrid adverse weather Conditions in one go. Specifically, our RAHC leverages a multi-head aggregation architecture to learn multiple degradation representation subspaces and then constrains the network to flexibly handle multiple hybrid adverse weather in a unified paradigm through a discrimination mechanism in the output space. Furthermore, we devise a reconstruction vectors aided scheme to provide auxiliary visual content cues for reconstruction, thus can comfortably cope with hybrid scenarios with insufficient remaining image constituents. Second, we construct a new dataset, termed HAC, for learning and benchmarking arbitrary Hybrid Adverse Conditions restoration. HAC contains 31 scenarios composed of an arbitrary combination of five common weather, with a total of ~316K adverse-weather/clean pairs. Extensive experiments yield superior results and establish new state-of-the-art results on both HAC and conventional datasets., Comment: In submission

In this paper, we will consider a class of continuous-time stochastic control systems with both unknown nonlinear structure and unknown disturbances, and investigate the capability of the classical proportional-integral-derivative(PID) controller in tracking time-varying reference signals. First, under some suitable conditions on system nonlinear functions, reference signals, and unknown disturbances, we will show that PID controllers can be designed to globally stabilize such systems and ensure the boundedness of the tracking error. Analytic design formulae for PID gain matrices are also provided, which only involve some prior knowledge of the partial derivatives of system structural nonlinear functions. Besides, it will be shown that the steady-state tracking error hinges on three critical factors: i) the change rate of reference signals and external disturbances; ii) the intensity of random noises; iii) the selection of PID gains, and can be made arbitrarily small by choosing PID gains suitably large. Finally, by introducing a desired transient process which is shaped from the reference signal, we will present a new PID tuning rule, which can guarantee both nice steady-state and superior transient control performances.

We perform weak lensing tomographic peak studies using the first-year shear data from Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) survey. The effective area used in our analyses after field selection, mask and boundary exclusions is $\sim 58 \deg^2$. The source galaxies are divided into low- and high-redshift bins with $0.2\le z_p\le0.85$ and $0.85\le z_p\le1.5$, respectively. We utilize our halo-based theoretical peak model including the projection effect of large-scale structures to derive cosmological constraints from the observed tomographic high peak abundances with the signal-to-noise ratio in the range of $\nu_{\rm N}=[3.5,5.5]$. These high peaks are closely associated with the lensing effects of massive clusters of galaxies. Thus the inclusion of their member galaxies in the shear catalog can lead to significant source clustering and dilute their lensing signals. We account for this systematic effect in our theoretical modelling. Additionally, the impacts of baryonic effects, galaxy intrinsic alignments, as well as residual uncertainties in shear and photometric redshift calibrations are also analyzed. Within the flat $\Lambda$CDM model, the derived constraint is $S_8=0.758_{-0.076}^{+0.033}$ and $0.768_{-0.057}^{+0.030}$ with the source clustering information measured from the two cluster catalogs, CAMIRA and WZL, respectively. The asymmetric uncertainties are due to the different degeneracy direction of $(\Omega_{\rm m}, \sigma_8)$ from high peak abundances comparing to that from the cosmic shear two-point correlations which give rise approximately the power index $\alpha=0.5$. Fitting to our constraints, we obtain $\alpha\approx 0.38$ and $\Sigma_8=0.772_{-0.032}^{+0.028}$ (CAMIRA) and $0.781_{-0.033}^{+0.028}$ (WZL). In comparison with the results from non-tomographic peak analyses, the $1\sigma$ uncertainties on $\Sigma_8$ are reduced by a factor of $\sim1.3$., Comment: 19 pages, 11 figures, accepted for publication in MNRAS

Real-Time Bidding (RTB) is an important mechanism in modern online advertising systems. Advertisers employ bidding strategies in RTB to optimize their advertising effects subject to various financial requirements, especially the return-on-investment (ROI) constraint. ROIs change non-monotonically during the sequential bidding process, and often induce a see-saw effect between constraint satisfaction and objective optimization. While some existing approaches show promising results in static or mildly changing ad markets, they fail to generalize to highly dynamic ad markets with ROI constraints, due to their inability to adaptively balance constraints and objectives amidst non-stationarity and partial observability. In this work, we specialize in ROI-Constrained Bidding in non-stationary markets. Based on a Partially Observable Constrained Markov Decision Process, our method exploits an indicator-augmented reward function free of extra trade-off parameters and develops a Curriculum-Guided Bayesian Reinforcement Learning (CBRL) framework to adaptively control the constraint-objective trade-off in non-stationary ad markets. Extensive experiments on a large-scale industrial dataset with two problem settings reveal that CBRL generalizes well in both in-distribution and out-of-distribution data regimes, and enjoys superior learning efficiency and stability., Comment: Accepted by SIGKDD 2022

We propose a new consistency test for the $\Lambda$CDM cosmology using baryonic acoustic oscillations (BAO) and redshift space distortion (RSD) measurements from galaxy redshift surveys. Specifically, we determine the peak position of $f\sigma_8(z)$ in redshift $z$ offered by a RSD measurement, and compare it to the one predicted by the BAO observables assuming a flat $\Lambda$CDM cosmology. We demonstrate this new test using the simulated data for the DESI galaxy survey, and argue that this test complements those using the background observables alone, and is less subject to systematics in the RSD analysis, compared to traditional methods using values of $f\sigma_8(z)$ directly., Comment: 9 pages, 6 figures; accepted to Research in Astronomy and Astrophysics (RAA)

To enlarge the perception range and reliability of individual autonomous vehicles, cooperative perception has been received much attention. However, considering the high volume of shared messages, limited bandwidth and computation resources in vehicular networks become bottlenecks. In this paper, we investigate how to balance the volume of shared messages and constrained resources in fog-based vehicular networks. To this end, we first characterize sum satisfaction of cooperative perception taking account of its spatial-temporal value and latency performance. Next, the sensing block message, communication resource block, and computation resource are jointly allocated to maximize the sum satisfaction of cooperative perception, while satisfying the maximum latency and sojourn time constraints of vehicles. Owing to its non-convexity, we decouple the original problem into two separate sub-problems and devise corresponding solutions. Simulation results demonstrate that our proposed scheme can effectively boost the sum satisfaction of cooperative perception compared with existing baselines., Comment: Accepted by WCSP 2021

Introducing blockchain into Federated Learning (FL) to build a trusted edge computing environment for transmission and learning has attracted widespread attention as a new decentralized learning pattern. However, traditional consensus mechanisms and architectures of blockchain systems face significant challenges in handling large-scale FL tasks, especially on Internet of Things (IoT) devices, due to their substantial resource consumption, limited transaction throughput, and complex communication requirements. To address these challenges, this paper proposes ChainFL, a novel two-layer blockchain-driven FL system. It splits the IoT network into multiple shards within the subchain layer, effectively reducing the scale of information exchange, and employs a Direct Acyclic Graph (DAG)-based mainchain as the mainchain layer, enabling parallel and asynchronous cross-shard validation. Furthermore, the FL procedure is customized to integrate deeply with blockchain technology, and a modified DAG consensus mechanism is designed to mitigate distortion caused by abnormal models. To provide a proof-of-concept implementation and evaluation, multiple subchains based on Hyperledger Fabric and a self-developed DAG-based mainchain are deployed. Extensive experiments demonstrate that ChainFL significantly surpasses conventional FL systems, showing up to a 14% improvement in training efficiency and a threefold increase in robustness., Comment: Accepted by IEEE Transactions on Network Science and Engineering