BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Lei Lei

Lei Lei contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Machine LearningArtificial IntelligenceComputer Visioneess.SPeess.SYInformation Theorymath.ITSystems and Controlastro-ph.GAastro-ph.IMastro-ph.SRDistributed, Parallel, and Cluster ComputingPerformancephysics.opticsRobotics

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
13works
0followers
15topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

13 published item(s)

preprint2026arXiv

DeformMaster: An Interactive Physics-Neural World Model for Deformable Objects from Videos

World models for deformable objects should recover not only geometry and appearance, but also underlying physical dynamics, interaction grounding, and material behavior. Learning such a model from real videos is challenging because deformable linear, planar, and volumetric objects evolve under high-dimensional deformation, noisy interactions, and complex material response. The model must therefore infer a physical state from visual observations, roll it forward under new interactions, and render the resulting dynamics with high visual fidelity. We present DeformMaster, a video-derived interactive physics--neural world model that turns real interaction videos into an online interactive model of deformable objects within a unified dynamics-and-appearance framework. DeformMaster preserves structured physical rollout while using a neural residual to compensate for unmodeled effects, grounds sparse hand motion as distributed compliant actuator for hand--continuum interaction, represents material response with spatially varying constitutive experts, and drives high-fidelity 4D appearance from the predicted physical evolution. Experiments on real-world deformable-object sequences demonstrate DeformMaster's ability to roll out future dynamics and render dynamic appearance, outperforming state-of-the-art baselines while supporting novel action rollout, material-parameter variation, and dynamic novel-view synthesis.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Physics-embedded neural computational electron microscopy for quantitative 4D nanometrology

The fusion of rigorous physical laws with flexible data-driven learning represents a new frontier in scientific simulation, yet bridging the gap between physical interpretability and computational efficiency remains a grand challenge. In electron microscopy, this divide limits the ability to quantify three-dimensional topography from two-dimensional projections, fundamentally constraining our understanding of nanoscale structure-function relationships. Here, we present a physics-embedded neural computational microscopy framework that achieves metrological three-dimensional reconstruction by deeply coupling a differentiable electron-optical forward model with deep learning. By introducing a Vision Field Transformer as a high-speed, differentiable surrogate for physical process analysis simulations, we establish an end-to-end, self-supervised optimization loop that enforces strict physical consistency with hardware geometry. This synergy enables single-shot, quantitative three-dimensional nanometrology with precision comparable to atomic force microscopy but at orders of magnitude higher throughput. Furthermore, we demonstrate the capability for four-dimensional (3D real space plus time) in situ characterization by tracking the dynamic evolution of surface nanostructure during copper redox, revealing hidden crystallographic kinetics invisible to conventional imaging. Our work not only redefines the limits of scanning electron microscopy but also establishes a generalizable archetype for solving ill-posed inverse problems across physical sciences, unlocking the full potential of simulation as a third pillar of discovery.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting

Neural rendering techniques, including NeRF and Gaussian Splatting (GS), rely on photometric consistency to produce high-quality reconstructions. However, in real-world scenarios, it is challenging to guarantee perfect photometric consistency in acquired images. Appearance codes have been widely used to address this issue, but their modeling capability is limited, as a single code is applied to the entire image. Recently, the bilateral grid was introduced to perform pixel-wise color mapping, but it is difficult to optimize and constrain effectively. In this paper, we propose a novel multi-scale bilateral grid that unifies appearance codes and bilateral grids. We demonstrate that this approach significantly improves geometric accuracy in dynamic, decoupled autonomous driving scene reconstruction, outperforming both appearance codes and bilateral grids. This is crucial for autonomous driving, where accurate geometry is important for obstacle avoidance and control. Our method shows strong results across four datasets: Waymo, NuScenes, Argoverse, and PandaSet. We further demonstrate that the improvement in geometry is driven by the multi-scale bilateral grid, which effectively reduces floaters caused by photometric inconsistency.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Meta Reinforcement Learning Approach for Predictive Autoscaling in the Cloud

Predictive autoscaling (autoscaling with workload forecasting) is an important mechanism that supports autonomous adjustment of computing resources in accordance with fluctuating workload demands in the Cloud. In recent works, Reinforcement Learning (RL) has been introduced as a promising approach to learn the resource management policies to guide the scaling actions under the dynamic and uncertain cloud environment. However, RL methods face the following challenges in steering predictive autoscaling, such as lack of accuracy in decision-making, inefficient sampling and significant variability in workload patterns that may cause policies to fail at test time. To this end, we propose an end-to-end predictive meta model-based RL algorithm, aiming to optimally allocate resource to maintain a stable CPU utilization level, which incorporates a specially-designed deep periodic workload prediction model as the input and embeds the Neural Process to guide the learning of the optimal scaling actions over numerous application services in the Cloud. Our algorithm not only ensures the predictability and accuracy of the scaling strategy, but also enables the scaling decisions to adapt to the changing workloads with high sample efficiency. Our method has achieved significant performance improvement compared to the existing algorithms and has been deployed online at Alipay, supporting the autoscaling of applications for the world-leading payment platform.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

AVDDPG: Federated reinforcement learning applied to autonomous platoon control

Since 2016 federated learning (FL) has been an evolving topic of discussion in the artificial intelligence (AI) research community. Applications of FL led to the development and study of federated reinforcement learning (FRL). Few works exist on the topic of FRL applied to autonomous vehicle (AV) platoons. In addition, most FRL works choose a single aggregation method (usually weight or gradient aggregation). We explore FRL's effectiveness as a means to improve AV platooning by designing and implementing an FRL framework atop a custom AV platoon environment. The application of FRL in AV platooning is studied under two scenarios: (1) Inter-platoon FRL (Inter-FRL) where FRL is applied to AVs across different platoons; (2) Intra-platoon FRL (Intra-FRL) where FRL is applied to AVs within a single platoon. Both Inter-FRL and Intra-FRL are applied to a custom AV platooning environment using both gradient and weight aggregation to observe the performance effects FRL can have on AV platoons relative to an AV platooning environment trained without FRL. It is concluded that Intra-FRL using weight aggregation (Intra-FRLWA) provides the best performance for controlling an AV platoon. In addition, we found that weight aggregation in FRL for AV platooning provides increases in performance relative to gradient aggregation. Finally, a performance analysis is conducted for Intra-FRLWA versus a platooning environment without FRL for platoons of length 3, 4 and 5 vehicles. It is concluded that Intra-FRLWA largely out-performs the platooning environment that is trained without FRL.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Deep Reinforcement Learning Aided Platoon Control Relying on V2X Information

The impact of Vehicle-to-Everything (V2X) communications on platoon control performance is investigated. Platoon control is essentially a sequential stochastic decision problem (SSDP), which can be solved by Deep Reinforcement Learning (DRL) to deal with both the control constraints and uncertainty in the platoon leading vehicle's behavior. In this context, the value of V2X communications for DRL-based platoon controllers is studied with an emphasis on the tradeoff between the gain of including exogenous information in the system state for reducing uncertainty and the performance erosion due to the curse-of-dimensionality. Our objective is to find the specific set of information that should be shared among the vehicles for the construction of the most appropriate state space. SSDP models are conceived for platoon control under different information topologies (IFT) by taking into account `just sufficient' information. Furthermore, theorems are established for comparing the performance of their optimal policies. In order to determine whether a piece of information should or should not be transmitted for improving the DRL-based control policy, we quantify its value by deriving the conditional KL divergence of the transition models. More meritorious information is given higher priority in transmission, since including it in the state space has a higher probability in offsetting the negative effect of having higher state dimensions. Finally, simulation results are provided to illustrate the theoretical analysis.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

An Ensemble Deep Convolutional Neural Network Model for Electricity Theft Detection in Smart Grids

Smart grids extremely rely on Information and Communications Technology (ICT) and smart meters to control and manage numerous parameters of the network. However, using these infrastructures make smart grids more vulnerable to cyber threats especially electricity theft. Electricity Theft Detection (EDT) algorithms are typically used for such purpose since this Non-Technical Loss (NTL) may lead to significant challenges in the power system. In this paper, an Ensemble Deep Convolutional Neural Network (EDCNN) algorithm for ETD in smart grids has been proposed. As the first layer of the model, a random under bagging technique is applied to deal with the imbalance data, and then Deep Convolutional Neural Networks (DCNN) are utilized on each subset. Finally, a voting system is embedded, in the last part. The evaluation results based on the Area Under Curve (AUC), precision, recall, f1-score, and accuracy verify the efficiency of the proposed method compared to the existing method in the literature.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Identifications of RR Lyrae stars and Quasars from the simulated data of Mephisto-W Survey

We have investigated the feasibilities and accuracies of the identifications of RR Lyrae stars and quasars from the simulated data of the Multi-channel Photometric Survey Telescope (Mephisto) W Survey. Based on the variable sources light curve libraries from the Sloan Digital Sky Survey (SDSS) Stripe 82 data and the observation history simulation from the Mephisto-W Survey Scheduler, we have simulated the $uvgriz$ multi-band light curves of RR Lyrae stars, quasars and other variable sources for the first year observation of Mephisto-W Survey. We have applied the ensemble machine learning algorithm Random Forest Classifier (RFC) to identify RR Lyrae stars and quasars, respectively. We build training and test samples and extract ~ 150 features from the simulated light curves and train two RFCs respectively for the RR Lyrae star and quasar classification. We find that, our RFCs are able to select the RR Lyrae stars and quasars with remarkably high precision and completeness, with $purity$ = 95.4 per cent and $completeness$ = 96.9 per cent for the RR Lyrae RFC and $purity$ = 91.4 per cent and $completeness$ = 90.2 per cent for the quasar RFC. We have also derived relative importances of the extracted features utilized to classify RR Lyrae stars and quasars.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Deep Reinforcement Learning for Autonomous Internet of Things: Model, Applications and Challenges

The Internet of Things (IoT) extends the Internet connectivity into billions of IoT devices around the world, where the IoT devices collect and share information to reflect status of the physical world. The Autonomous Control System (ACS), on the other hand, performs control functions on the physical systems without external intervention over an extended period of time. The integration of IoT and ACS results in a new concept - autonomous IoT (AIoT). The sensors collect information on the system status, based on which the intelligent agents in the IoT devices as well as the Edge/Fog/Cloud servers make control decisions for the actuators to react. In order to achieve autonomy, a promising method is for the intelligent agents to leverage the techniques in the field of artificial intelligence, especially reinforcement learning (RL) and deep reinforcement learning (DRL) for decision making. In this paper, we first provide a tutorial of DRL, and then propose a general model for the applications of RL/DRL in AIoT. Next, a comprehensive survey of the state-of-art research on DRL for AIoT is presented, where the existing works are classified and summarized under the umbrella of the proposed general DRL model. Finally, the challenges and open issues for future research are identified.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Energy Minimization in UAV-Aided Networks: Actor-Critic Learning for Constrained Scheduling Optimization

In unmanned aerial vehicle (UAV) applications, the UAV's limited energy supply and storage have triggered the development of intelligent energy-conserving scheduling solutions. In this paper, we investigate energy minimization for UAV-aided communication networks by jointly optimizing data-transmission scheduling and UAV hovering time. The formulated problem is combinatorial and non-convex with bilinear constraints. To tackle the problem, firstly, we provide an optimal relax-and-approximate solution and develop a near-optimal algorithm. Both the proposed solutions are served as offline performance benchmarks but might not be suitable for online operation. To this end, we develop a solution from a deep reinforcement learning (DRL) aspect. The conventional RL/DRL, e.g., deep Q-learning, however, is limited in dealing with two main issues in constrained combinatorial optimization, i.e., exponentially increasing action space and infeasible actions. The novelty of solution development lies in handling these two issues. To address the former, we propose an actor-critic-based deep stochastic online scheduling (AC-DSOS) algorithm and develop a set of approaches to confine the action space. For the latter, we design a tailored reward function to guarantee the solution feasibility. Numerical results show that, by consuming equal magnitude of time, AC-DSOS is able to provide feasible solutions and saves 29.94% energy compared with a conventional deep actor-critic method. Compared to the developed near-optimal algorithm, AC-DSOS consumes around 10% higher energy but reduces the computational time from minute-level to millisecond-level.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

NOMA-Enabled Multi-Beam Satellite Systems: Joint Optimization to Overcome Offered-Requested Data Mismatches

Non-Orthogonal Multiple Access (NOMA) has potentials to improve the performance of multi-beam satellite systems. The performance optimization in satellite-NOMA systems can be different from that in terrestrial-NOMA systems, e.g., considering distinctive channel models, performance metrics, power constraints, and limited flexibility in resource management. In this paper, we adopt a metric, Offered Capacity to requested Traffic Ratio (OCTR), to measure the requested-offered data (or rate) mismatch in multi-beam satellite systems. In the considered system, NOMA is applied to mitigate intra-beam interference while precoding is implemented to reduce inter-beam interference. We jointly optimize power, decoding orders, and terminal-timeslot assignment to improve the max-min fairness of OCTR. The problem is inherently difficult due to the presence of combinatorial and non-convex aspects. We first fix the terminal-timeslot assignment, and develop an optimal fast-convergence algorithmic framework based on Perron-Frobenius theory (PF) for the remaining joint power-allocation and decoding-order optimization problem. Under this framework, we propose a heuristic algorithm for the original problem, which iteratively updates the terminal-timeslot assignment and improves the overall OCTR performance. Numerical results verify that max-min OCTR is a suitable metric to address the mismatch issue, and is able to improve the fairness among terminals. In average, the proposed algorithm improves the max-min OCTR by 40.2% over Orthogonal Multiple Access (OMA).

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Performance Modeling and Analysis of a Hyperledger-based System Using GSPN

As a highly scalable permissioned blockchain platform, Hyperledger Fabric supports a wide range of industry use cases ranging from governance to finance. In this paper, we propose a model to analyze the performance of a Hyperledgerbased system by using Generalised Stochastic Petri Nets (GSPN). This model decomposes a transaction flow into multiple phases and provides a simulation-based approach to obtain the system latency and throughput with a specific arrival rate. Based on this model, we analyze the impact of different configurations of ordering service on system performance to find out the bottleneck. Moreover, a mathematical configuration selection approach is proposed to determine the best configuration which can maximize the system throughput. Finally, extensive experiments are performed on a running system to validate the proposed model and approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Towards Power-Efficient Aerial Communications via Dynamic Multi-UAV Cooperation

Aerial base stations (BSs) attached to unmanned aerial vehicles (UAVs) constitute a new paradigm for next-generation cellular communications. However, the flight range and communication capacity of aerial BSs are usually limited due to the UAVs' size, weight, and power (SWAP) constraints. To address this challenge, in this paper, we consider dynamic cooperative transmission among multiple aerial BSs for power-efficient aerial communications. Thereby, a central controller intelligently selects the aerial BSs navigating in the air for cooperation. Consequently, the large virtual array of moving antennas formed by the cooperating aerial BSs can be exploited for low-power information transmission and navigation, taking into account the channel conditions, energy availability, and user demands. Considering both the fronthauling and the data transmission links, we jointly optimize the trajectories, cooperation decisions, and transmit beamformers of the aerial BSs for minimization of the weighted sum of the power consumptions required by all BSs. Since obtaining the global optimal solution of the formulated problem is difficult, we propose a low-complexity iterative algorithm that can efficiently find a Karush-Kuhn-Tucker (KKT) solution to the problem. Simulation results show that, compared with several baseline schemes, dynamic multi-UAV cooperation can significantly reduce the communication and navigation powers of the UAVs to overcome the SWAP limitations, while requiring only a small increase of the transmit power over the fronthauling links.

0 citations0 reviewsNo code linkedOpen access