arXiv subjects

Hong Xie

Publications and source records attributed to Hong Xie.

At least 73 records · Page 4Linked to original sources

arXiv2023

Uncertainty-Aware Instance Reweighting for Off-Policy Learning

Off-policy learning, referring to the procedure of policy optimization with access only to logged feedback data, has shown importance in various real-world applications, such as search engines, recommender systems, and etc. While the ground-truth logging policy, which generates the logged data, is usually unknown, previous work simply takes its estimated value in off-policy learning, ignoring both high bias and high variance resulted from such an estimator, especially on samples with small and inaccurately estimated logging probabilities. In this work, we explicitly model the uncertainty in the estimated logging policy and propose a Uncertainty-aware Inverse Propensity Score estimator (UIPS) for improved off-policy learning, with a theoretical convergence guarantee. Experiment results on synthetic and three real-world recommendation datasets demonstrate the advantageous sample efficiency of the proposed UIPS estimator against an extensive list of state-of-the-art baselines.

cs.LG↗

arXiv2023

Interactive Graph Convolutional Filtering

Interactive Recommender Systems (IRS) have been increasingly used in various domains, including personalized article recommendation, social media, and online advertising. However, IRS faces significant challenges in providing accurate recommendations under limited observations, especially in the context of interactive collaborative filtering. These problems are exacerbated by the cold start problem and data sparsity problem. Existing Multi-Armed Bandit methods, despite their carefully designed exploration strategies, often struggle to provide satisfactory results in the early stages due to the lack of interaction data. Furthermore, these methods are computationally intractable when applied to non-linear models, limiting their applicability. To address these challenges, we propose a novel method, the Interactive Graph Convolutional Filtering model. Our proposed method extends interactive collaborative filtering into the graph model to enhance the performance of collaborative filtering between users and items. We incorporate variational inference techniques to overcome the computational hurdles posed by non-linear models. Furthermore, we employ Bayesian meta-learning methods to effectively address the cold-start problem and derive theoretical regret bounds for our proposed method, ensuring a robust performance guarantee. Extensive experimental results on three real-world datasets validate our method and demonstrate its superiority over existing baselines.

cs.IR↗

arXiv2023

What do halo CMEs tell us about solar cycle 25?

It is known that the weak state of the heliosphere due to diminished solar activity in cycle 24 back-reacted on coronal mass ejections (CMEs) to make them appear wider for a given speed. One of the consequences of the weak state of the heliosphere is that more CMEs appear as halo CMEs (HCMEs), and halos are formed at shorter heliocentric distances. Current predictions for the strength of solar cycle (SC) 25 range from half to twice the strength of SC 24. We compare the HCME occurrence rate and other properties during the rise phase of cycles 23, 24, and 25 to weigh in on the strength of SC 25. We find that HCME and solar wind properties in SC 25 are intermediate between SCs 23 and 24, but closer to SC 24. The HCME occurrence rate, normalized to the sunspot number, is higher in SCs 24 and 25 than in SC 23. The solar wind total pressure in SC 25 is ~35% smaller than that in SC 23. Furthermore, the occurrence rates of high-energy solar energetic particle events and intense geomagnetic storms are well below the corresponding values in SC 23, but similar to those in SC 24. We conclude that cycle 25 is likely to be similar to or slightly stronger than cycle 24, in agreement with polar-field precursor methods for cycle 25 prediction

astro-ph.SR↗

arXiv2023

The Solar Cause of the 2022 February 3 Geomagnetic Storm that Led to the Demise of the Starlink Satellites

We report on the solar source of the 2022 February 3 geomagnetic storm of moderate strength that contributed to the loss of 39 Starlink satellites. The geomagnetic storm was caused by the 2022 January 29 halo coronal mass ejection (CME) that was of moderate speed (about 690 km/s) originating from NOAA active region 12936 located in the northeast quadrant (N18E06) of the Sun. The eruption was marked by an M1.1 flare, which started at 22:45 UT, peaked at 23:32 UT on January 29 and ended at 00:24 UT the next day. The CME ended up as a shock-driving magnetic cloud (MC) observed at Sun-Earth L1 and at STEREO-Ahead (STA) located about 34 deg behind Earth. The geomagnetic storm was caused by a strong southward component of the MC that was boosted by a high speed solar wind stream behind the MC. Even though Earth and STA were separated by only about 34 deg, the MC appeared quite different at Earth and L1. One possibility is that the MC was writhed reflecting the curved neutral line at the Sun. In-situ observations suggest that the MC was heading closer to STA than to Earth because of the earlier arrival at STA. However, the shock arrived at STA and Earth around the same time, suggesting a weaker shock at Earth due to flank passage.

astro-ph.SR↗

arXiv2023

Interactive Log Parsing via Light-weight User Feedback

Template mining is one of the foundational tasks to support log analysis, which supports the diagnosis and troubleshooting of large scale Web applications. This paper develops a human-in-the-loop template mining framework to support interactive log analysis, which is highly desirable in real-world diagnosis or troubleshooting of Web applications but yet previous template mining algorithms fails to support it. We formulate three types of light-weight user feedbacks and based on them we design three atomic human-in-the-loop template mining algorithms. We derive mild conditions under which the outputs of our proposed algorithms are provably correct. We also derive upper bounds on the computational complexity and query complexity of each algorithm. We demonstrate the versatility of our proposed algorithms by combining them to improve the template mining accuracy of five representative algorithms over sixteen widely used benchmark datasets.

cs.AI↗

arXiv2023

Autonomous Exploration Method for Fast Unknown Environment Mapping by Using UAV Equipped with Limited FOV Sensor

Autonomous exploration is one of the important parts to achieve the fast autonomous mapping and target search. However, most of the existing methods are facing low-efficiency problems caused by low-quality trajectory or back-and-forth maneuvers. To improve the exploration efficiency in unknown environments, a fast autonomous exploration planner (FAEP) is proposed in this paper. Different from existing methods, we firstly design a novel frontiers exploration sequence generation method to obtain a more reasonable exploration path, which considers not only the flight-level but frontier-level factors in the asymmetric traveling salesman problem (ATSP). Then, according to the exploration sequence and the distribution of frontiers, an adaptive yaw planning method is proposed to cover more frontiers by yaw change during an exploration journey. In addition, to increase the speed and fluency of flight, a dynamic replanning strategy is also adopted. We present sufficient comparison and evaluation experiments in simulation environments. Experimental results show the proposed exploration planner has better performance in terms of flight time and flight distance compared to typical and state-of-the-art methods. Moreover, the effectiveness of the proposed method is further evaluated in real-world environments.

cs.RO↗

arXiv2022

Multiple-Play Stochastic Bandits with Shareable Finite-Capacity Arms

We generalize the multiple-play multi-armed bandits (MP-MAB) problem with a shareable arm setting, in which several plays can share the same arm. Furthermore, each shareable arm has a finite reward capacity and a ''per-load'' reward distribution, both of which are unknown to the learner. The reward from a shareable arm is load-dependent, which is the "per-load" reward multiplying either the number of plays pulling the arm, or its reward capacity when the number of plays exceeds the capacity limit. When the "per-load" reward follows a Gaussian distribution, we prove a sample complexity lower bound of learning the capacity from load-dependent rewards and also a regret lower bound of this new MP-MAB problem. We devise a capacity estimator whose sample complexity upper bound matches the lower bound in terms of reward means and capacities. We also propose an online learning algorithm to address the problem and prove its regret upper bound. This regret upper bound's first term is the same as regret lower bound's, and its second and third terms also evidently correspond to lower bound's. Extensive experiments validate our algorithm's performance and also its gain in 5G & 4G base station selection.

cs.LG↗

arXiv2022

Can Type III Radio Storms be a Source of Seed Particles to Shock Acceleration?

An intense type III radio storm has been disrupted by a fast halo coronal mass ejection (CME) on 2000 April 4. The CME is also associated with a large solar energetic particle (SEP) event. The storm recovers after about10 hrs. We identified another CME that occurs on 2003 November 11 with similar CME properties but there is no type III storm in progress. The 2003 November 11 CME is also not associated with an SEP event above the background (less than 2 pfu), whereas the one with type III storm has an intense SEP event (about 56 pfu). One of the factors affecting the intensity of SEP events is the presence of seed particles that are accelerated by CME-driven shocks. We suggest that the type III storm source, which accelerates electrons to produce the storm, also accelerates ions that serve as seed particles to the CME shock.

astro-ph.SR↗

arXiv2022

The Relation between Type III Radio Storms and CIR Energetic Particles

We report on a study that compares energetic particle fluxes in corotating interaction regions (CIRs) associated with type III radio storm with those in nonstorm CIRs. In a case study, we compare the CIR particle events on 2010October 21 and 2005 November 2. The two events have similar solar and solar wind circumstances, except that the former is associated with a type III radio storm and has a higher CIR particle flux and fluence. We also perform a statistical study, which shows that the proton and electron fluences are higher in the storm associated CIRs by factor of about 6 and 8, respectively than those in the storm-free CIRs.

astro-ph.SR↗

arXiv2022

A New Outlier Removal Strategy Based on Reliability of Correspondence Graph for Fast Point Cloud Registration

Registration is a basic yet crucial task in point cloud processing. In correspondence-based point cloud registration, matching correspondences by point feature techniques may lead to an extremely high outlier ratio. Current methods still suffer from low efficiency, accuracy, and recall rate. We use a simple and intuitive method to describe the 6-DOF (degree of freedom) curtailment process in point cloud registration and propose an outlier removal strategy based on the reliability of the correspondence graph. The method constructs the corresponding graph according to the given correspondences and designs the concept of the reliability degree of the graph node for optimal candidate selection and the reliability degree of the graph edge to obtain the global maximum consensus set. The presented method could achieve fast and accurate outliers removal along with gradual aligning parameters estimation. Extensive experiments on simulations and challenging real-world datasets demonstrate that the proposed method can still perform effective point cloud registration even the correspondence outlier ratio is over 99%, and the efficiency is better than the state-of-the-art. Code is available at https://github.com/WPC-WHU/GROR.

cs.CV↗

arXiv2022

LPC-AD: Fast and Accurate Multivariate Time Series Anomaly Detection via Latent Predictive Coding

This paper proposes LPC-AD, a fast and accurate multivariate time series (MTS) anomaly detection method. LPC-AD is motivated by the ever-increasing needs for fast and accurate MTS anomaly detection methods to support fast troubleshooting in cloud computing, micro-service systems, etc. LPC-AD is fast in the sense that its reduces the training time by as high as 38.2% compared to the state-of-the-art (SOTA) deep learning methods that focus on training speed. LPC-AD is accurate in the sense that it improves the detection accuracy by as high as 18.9% compared to SOTA sophisticated deep learning methods that focus on enhancing detection accuracy. Methodologically, LPC-AD contributes a generic architecture LPC-Reconstruct for one to attain different trade-offs between training speed and detection accuracy. More specifically, LPC-Reconstruct is built on ideas from autoencoder for reducing redundancy in time series, latent predictive coding for capturing temporal dependence in MTS, and randomized perturbation for avoiding overfitting of anomalous dependence in the training data. We present simple instantiations of LPC-Reconstruct to attain fast training speed, where we propose a simple randomized perturbation method. The superior performance of LPC-AD over SOTA methods is validated by extensive experiments on four large real-world datasets. Experiment results also show the necessity and benefit of each component of the LPC-Reconstruct architecture and that LPC-AD is robust to hyper parameters.

cs.LG↗

arXiv2022

Multi-Player Multi-Armed Bandits with Finite Shareable Resources Arms: Learning Algorithms & Applications

Multi-player multi-armed bandits (MMAB) study how decentralized players cooperatively play the same multi-armed bandit so as to maximize their total cumulative rewards. Existing MMAB models mostly assume when more than one player pulls the same arm, they either have a collision and obtain zero rewards, or have no collision and gain independent rewards, both of which are usually too restrictive in practical scenarios. In this paper, we propose an MMAB with shareable resources as an extension to the collision and non-collision settings. Each shareable arm has finite shareable resources and a "per-load" reward random variable, both of which are unknown to players. The reward from a shareable arm is equal to the "per-load" reward multiplied by the minimum between the number of players pulling the arm and the arm's maximal shareable resources. We consider two types of feedback: sharing demand information (SDI) and sharing demand awareness (SDA), each of which provides different signals of resource sharing. We design the DPE-SDI and SIC-SDA algorithms to address the shareable arm problem under these two cases of feedback respectively and prove that both algorithms have logarithmic regrets that are tight in the number of rounds. We conduct simulations to validate both algorithms' performance and show their utilities in wireless networking and edge computing.

cs.LG↗

arXiv2022

FAEP: Fast Autonomous Exploration Planner for UAV Equipped with Limited FOV Sensor

Autonomous exploration is one of the important parts to achieve the autonomous operation of Unmanned Aerial Vehicles (UAVs). To improve the efficiency of the exploration process, a fast and autonomous exploration planner (FAEP) is proposed in this paper. We firstly design a novel frontiers exploration sequence generation method to obtain a more reasonable exploration path, which considers not only the flight-level but frontier-level factors into TSP. According to the exploration sequence and the distribution of frontiers, a two-stage heading planning strategy is proposed to cover more frontiers by heading change during an exploration journey. To improve the stability of path searching, a guided kinodynamic path searching based on a guiding path is devised. In addition, a dynamic start point selection method for replanning is also adopted to increase the fluency of flight. We present sufficient benchmark and real-world experiments. Experimental results show the superiority of the proposed exploration planner compared with typical and state-of-the-art methods.

cs.RO↗

arXiv2022

Proposal for a Bell test in cavity optomagnonics

We present a proposal to test Bell inequality in the emerging field of cavity optomagnonics, where a sphere of ferromagnetic crystal supports two optical whispering gallery modes and one magnon mode. The two optical modes are driven by two laser pulses, respectively. Entanglement between magnon mode and one of the two optical modes will be generated by the first pulse, and the state of magnon mode is subsequently mapped into another optical mode via the second pulse. Hence correlated photon-photon pairs is created out of the cavity. A Bell test can be implemented using these pairs, which enables us to test local hidden-variable models at macroscopic optomagnonical system. Our results show that a significant violation of Bell inequality can be obtained in the experimentally relevant weak-coupling regime. The violation of Bell inequality not only verifies the entanglement between magnons and photons, but also implies that cavity optomagnonics is a promising platform for quantum information processing tasks.

quant-ph↗

arXiv2022

Direct First PSP Observation of the Interaction of Two Successive Interplanetary Coronal Mass Ejections in November 2020

We investigate the effects of the evolutionary processes in the internal magnetic structure of two interplanetary coronal mass ejections (ICMEs) detected in situ between 2020 November 29 and December 1 by Parker Solar Probe (PSP). The sources of the ICMEs were observed remotely at the Sun in EUV and subsequently tracked to their coronal counterparts in white light. This period is of particular interest to the community since it has been identified as the first widespread solar energetic particle event of Solar Cycle 25. The distribution of various solar and heliospheric-dedicated spacecraft throughout the inner heliosphere during PSP observations of these large-scale magnetic structures enables a comprehensive analysis of the internal evolution and topology of such structures. By assembling different models and techniques, we identify the signatures of interaction between the two consecutive ICMEs and the implications for their internal structure. We use multispacecraft observations in combination with a remote-sensing forward modeling technique, numerical propagation models, and in-situ reconstruction techniques. The outcome, from the full reconciliations, demonstrates that the two CMEs are interacting in the vicinity of PSP. Thus, we identify the in-situ observations based on the physical processes that are associated with the interaction and collision of both CMEs. We also expand the flux rope modeling and in-situ reconstruction technique to incorporate the aging and expansion effects in a distorted internal magnetic structure and explore the implications of both effects in the magnetic configuration of the ICMEs.

astro-ph.SR↗

arXiv2022

Solar Activity and Space Weather

After providing an overview of solar activity as measured by the sunspot number (SSN) and space weather events during solar cycles (SCs) 21-24, we focus on the weak solar activity in SC 24. The weak solar activity reduces the number of energetic eruptions from the Sun and hence the number of space weather events. The speeds of coronal mass ejections (CMEs), interplanetary (IP) shocks, and the background solar wind all declined in SC 24. One of the main heliospheric consequences of weak solar activity is the reduced total (magnetic + gas) pressure, magnetic field strength, and Alfvén speed. There are three groups of phenomena that decline to different degrees in SC 24 relative to the corresponding ones in SC 23: (i) those that decline more than SSN does, (ii) those that decline like SSN, and (iii) those that decline less than SSN does. The decrease in the number of severe space weather events such as high-energy solar energetic particle (SEP) events and intense geomagnetic storms is deeper than the decline in SSN. CMEs expand anomalously and hence their magnetic content is diluted resulting in weaker geomagnetic storms. The reduction in the number of intense geomagnetic storms caused by corotating interaction regions is also drastic. The diminished heliospheric magnetic field in SC 24 reduces the efficiency of particle acceleration, resulting in fewer high-energy SEP events. The numbers of IP type II radio bursts, IP socks, and high-intensity energetic storm particle events closely follow the number of fast and wide CMEs (and approximately SSN). The number of halo CMEs in SC 24 declines less than SSN does, mainly due to the weak heliospheric state. Phenomena such as IP CMEs and magnetic clouds related to frontside halos also do not decline significantly. The mild space weather is likely to continue in SC 25, whose strength has been predicted to be not too different from that of SC 24.

astro-ph.SR↗

arXiv2021

Does Explicit Prediction Matter in Deep Reinforcement Learning-Based Energy Management?

As a model-free optimization and decision-making method, deep reinforcement learning (DRL) has been widely applied to the filed of energy management in energy Internet. While, some DRL-based energy management schemes also incorporate the prediction module used by the traditional model-based methods, which seems to be unnecessary and even adverse. In this work, we implement the standard energy management scheme with prediction using supervised learning and DRL, and the counterpart without prediction using end-to-end DRL. Then, these two schemes are compared in the unified energy management framework. The simulation results demonstrate that the energy management scheme without prediction is superior over the scheme with prediction. This work intends to rectify the misuse of DRL methods in the field of energy management.

eess.SY↗

arXiv2021

Pricing Social Visibility Service in Online Social Networks: Modeling and Algorithms

Online social networks (OSNs) such as YoutTube, Instagram, Twitter, Facebook, etc., serve as important platforms for users to share their information or content to friends or followers. Oftentimes, users want to enhance their social visibility, as it can make their contents, i.e., opinions, videos, pictures, etc., attract attention from more users, which in turn may bring higher commercial benefit to them. Motivated by this, we propose a mechanism, where the OSN operator provides a "social visibility boosting service" to incentivize "transactions" between requesters (users who seek to enhance their social visibility via adding new "neighbors") and suppliers (users who are willing to be added as a new "neighbor" of any requester when certain "rewards" is provided). We design a posted pricing scheme for the OSN provider to charge the requesters who use such boosting service, and reward the suppliers who contribute to such boosting service. The OSN operator keeps a fraction of the payment from requesters and distributes the remaining part to participating suppliers "fairly" via the Shapley value. The objective of the OSN provider is to select the price and supplier set to maximize the total amount of revenue under the budget constraint of requesters. We first show that the revenue maximization problem is not simpler than an NP-hard problem. We then decomposed it into two sub-routines, where one focuses on selecting the optimal set of suppliers, and the other one focuses on selecting the optimal price. We prove the hardness of each sub-routine, and eventually design a computationally efficient approximation algorithm to solve the revenue maximization problem with provable theoretical guarantee on the revenue gap. We conduct extensive experiments on four public datasets to validate the superior performance of our proposed algorithms.

cs.SI↗