arXiv subjects

Ke Li

Publications and source records attributed to Ke Li.

At least 289 records · Page 16Linked to original sources

arXiv2023

Magic NeRF Lens: Interactive Fusion of Neural Radiance Fields for Virtual Facility Inspection

Large industrial facilities such as particle accelerators and nuclear power plants are critical infrastructures for scientific research and industrial processes. These facilities are complex systems that not only require regular maintenance and upgrades but are often inaccessible to humans due to various safety hazards. Therefore, a virtual reality (VR) system that can quickly replicate real-world remote environments to provide users with a high level of spatial and situational awareness is crucial for facility maintenance planning. However, the exact 3D shapes of these facilities are often too complex to be accurately modeled with geometric primitives through the traditional rasterization pipeline. In this work, we develop Magic NeRF Lens, an interactive framework to support facility inspection in immersive VR using neural radiance fields (NeRF) and volumetric rendering. We introduce a novel data fusion approach that combines the complementary strengths of volumetric rendering and geometric rasterization, allowing a NeRF model to be merged with other conventional 3D data, such as a computer-aided design model. We develop two novel 3D magic lens effects to optimize NeRF rendering by exploiting the properties of human vision and context-aware visualization. We demonstrate the high usability of our framework and methods through a technical benchmark, a visual search user study, and expert reviews. In addition, the source code of our VR NeRF framework is made publicly available for future research and development.

cs.GR↗

arXiv2023

Model-Assisted Probabilistic Safe Adaptive Control With Meta-Bayesian Learning

Breaking safety constraints in control systems can lead to potential risks, resulting in unexpected costs or catastrophic damage. Nevertheless, uncertainty is ubiquitous, even among similar tasks. In this paper, we develop a novel adaptive safe control framework that integrates meta learning, Bayesian models, and control barrier function (CBF) method. Specifically, with the help of CBF method, we learn the inherent and external uncertainties by a unified adaptive Bayesian linear regression (ABLR) model, which consists of a forward neural network (NN) and a Bayesian output layer. Meta learning techniques are leveraged to pre-train the NN weights and priors of the ABLR model using data collected from historical similar tasks. For a new control task, we refine the meta-learned models using a few samples, and introduce pessimistic confidence bounds into CBF constraints to ensure safe control. Moreover, we provide theoretical criteria to guarantee probabilistic safety during the control processes. To validate our approach, we conduct comparative experiments in various obstacle avoidance scenarios. The results demonstrate that our algorithm significantly improves the Bayesian model-based CBF method, and is capable for efficient safe exploration even with multiple uncertain constraints.

eess.SY↗

arXiv2023

Defend Data Poisoning Attacks on Voice Authentication

With the advances in deep learning, speaker verification has achieved very high accuracy and is gaining popularity as a type of biometric authentication option in many scenes of our daily life, especially the growing market of web services. Compared to traditional passwords, "vocal passwords" are much more convenient as they relieve people from memorizing different passwords. However, new machine learning attacks are putting these voice authentication systems at risk. Without a strong security guarantee, attackers could access legitimate users' web accounts by fooling the deep neural network (DNN) based voice recognition models. In this paper, we demonstrate an easy-to-implement data poisoning attack to the voice authentication system, which can hardly be captured by existing defense mechanisms. Thus, we propose a more robust defense method, called Guardian, which is a convolutional neural network-based discriminator. The Guardian discriminator integrates a series of novel techniques including bias reduction, input augmentation, and ensemble learning. Our approach is able to distinguish about 95% of attacked accounts from normal accounts, which is much more effective than existing approaches with only 60% accuracy.

cs.CR↗

arXiv2023

MEMD-ABSA: A Multi-Element Multi-Domain Dataset for Aspect-Based Sentiment Analysis

Aspect-based sentiment analysis is a long-standing research interest in the field of opinion mining, and in recent years, researchers have gradually shifted their focus from simple ABSA subtasks to end-to-end multi-element ABSA tasks. However, the datasets currently used in the research are limited to individual elements of specific tasks, usually focusing on in-domain settings, ignoring implicit aspects and opinions, and with a small data scale. To address these issues, we propose a large-scale Multi-Element Multi-Domain dataset (MEMD) that covers the four elements across five domains, including nearly 20,000 review sentences and 30,000 quadruples annotated with explicit and implicit aspects and opinions for ABSA research. Meanwhile, we evaluate generative and non-generative baselines on multiple ABSA subtasks under the open domain setting, and the results show that open domain ABSA as well as mining implicit aspects and opinions remain ongoing challenges to be addressed. The datasets are publicly released at \url{https://github.com/NUSTM/MEMD-ABSA}.

cs.CL↗

arXiv2023

Learning from Visual Observation via Offline Pretrained State-to-Go Transformer

Learning from visual observation (LfVO), aiming at recovering policies from only visual observation data, is promising yet a challenging problem. Existing LfVO approaches either only adopt inefficient online learning schemes or require additional task-specific information like goal states, making them not suited for open-ended tasks. To address these issues, we propose a two-stage framework for learning from visual observation. In the first stage, we introduce and pretrain State-to-Go (STG) Transformer offline to predict and differentiate latent transitions of demonstrations. Subsequently, in the second stage, the STG Transformer provides intrinsic rewards for downstream reinforcement learning tasks where an agent learns merely from intrinsic rewards. Empirical results on Atari and Minecraft show that our proposed method outperforms baselines and in some tasks even achieves performance comparable to the policy learned from environmental rewards. These results shed light on the potential of utilizing video-only data to solve difficult visual reinforcement learning tasks rather than relying on complete offline datasets containing states, actions, and rewards. The project's website and code can be found at https://sites.google.com/view/stgtransformer.

cs.LG↗

arXiv2023

EduChain: A Blockchain-based Education Data Management System

The predominant centralized paradigm in educational data management currently suffers from several critical issues such as vulnerability to malicious tampering, a high prevalence of diploma counterfeiting, and the onerous cost of certificate authentication. Decentralized blockchain technology, with its cutting-edge capabilities, presents a viable solution to these pervasive problems. In this paper, we illuminate the inherent limitations of existing centralized systems and introduce EduChain, a novel heterogeneous blockchain-based system for managing educational data. EduChain uniquely harnesses the strengths of both private and consortium blockchains, offering an unprecedented level of security and efficiency. In addition, we propose a robust mechanism for performing database consistency checks and error tracing. This is achieved through the implementation of a secondary consensus, employing the pt-table-checksum tool. This approach effectively addresses the prevalent issue of database mismatches. Our system demonstrates superior performance in key areas such as information verification, error traceback, and data security, thereby significantly improving the integrity and trustworthiness of educational data management. Through EduChain, we offer a powerful solution for future advancements in secure and efficient educational data management.

cs.CR↗

arXiv2023

Tight Exponential Analysis for Smoothing the Max-Relative Entropy and for Quantum Privacy Amplification

The max-relative entropy together with its smoothed version is a basic tool in quantum information theory. In this paper, we derive the exact exponent for the asymptotic decay of the small modification of the quantum state in smoothing the max-relative entropy based on purified distance. We then apply this result to the problem of privacy amplification against quantum side information, and we obtain an upper bound for the exponent of the asymptotic decreasing of the insecurity, measured using either purified distance or relative entropy. Our upper bound complements the earlier lower bound established by Hayashi, and the two bounds match when the rate of randomness extraction is above a critical value. Thus, for the case of high rate, we have determined the exact security exponent. Following this, we give examples and show that in the low-rate case, neither the upper bound nor the lower bound is tight in general. This exhibits a picture similar to that of the error exponent in channel coding. Lastly, we investigate the asymptotics of equivocation and its exponent under the security measure using the sandwiched Rényi divergence of order $s\in (1,2]$, which has not been addressed previously in the quantum setting.

quant-ph↗

arXiv2023

Integrating Multiple Sources Knowledge for Class Asymmetry Domain Adaptation Segmentation of Remote Sensing Images

In the existing unsupervised domain adaptation (UDA) methods for remote sensing images (RSIs) semantic segmentation, class symmetry is an widely followed ideal assumption, where the source and target RSIs have exactly the same class space. In practice, however, it is often very difficult to find a source RSI with exactly the same classes as the target RSI. More commonly, there are multiple source RSIs available. To this end, a novel class asymmetry RSIs domain adaptation method with multiple sources is proposed in this paper, which consists of four key components. Firstly, a multi-branch segmentation network is built to learn an expert for each source RSI. Secondly, a novel collaborative learning method with the cross-domain mixing strategy is proposed, to supplement the class information for each source while achieving the domain adaptation of each source-target pair. Thirdly, a pseudo-label generation strategy is proposed to effectively combine strengths of different experts, which can be flexibly applied to two cases where the source class union is equal to or includes the target class set. Fourthly, a multiview-enhanced knowledge integration module is developed for the high-level knowledge routing and transfer from multiple domains to target predictions.

cs.CV↗

arXiv2023

SketchXAI: A First Look at Explainability for Human Sketches

This paper, for the very first time, introduces human sketches to the landscape of XAI (Explainable Artificial Intelligence). We argue that sketch as a ``human-centred'' data form, represents a natural interface to study explainability. We focus on cultivating sketch-specific explainability designs. This starts by identifying strokes as a unique building block that offers a degree of flexibility in object construction and manipulation impossible in photos. Following this, we design a simple explainability-friendly sketch encoder that accommodates the intrinsic properties of strokes: shape, location, and order. We then move on to define the first ever XAI task for sketch, that of stroke location inversion SLI. Just as we have heat maps for photos, and correlation matrices for text, SLI offers an explainability angle to sketch in terms of asking a network how well it can recover stroke locations of an unseen sketch. We offer qualitative results for readers to interpret as snapshots of the SLI process in the paper, and as GIFs on the project page. A minor but interesting note is that thanks to its sketch-specific design, our sketch encoder also yields the best sketch recognition accuracy to date while having the smallest number of parameters. The code is available at \url{https://sketchxai.github.io}.

cs.CV↗

arXiv2023

Diffusion models with location-scale noise

Diffusion Models (DMs) are powerful generative models that add Gaussian noise to the data and learn to remove it. We wanted to determine which noise distribution (Gaussian or non-Gaussian) led to better generated data in DMs. Since DMs do not work by design with non-Gaussian noise, we built a framework that allows reversing a diffusion process with non-Gaussian location-scale noise. We use that framework to show that the Gaussian distribution performs the best over a wide range of other distributions (Laplace, Uniform, t, Generalized-Gaussian).

cs.LG↗

arXiv2023

ClusterFusion: Real-time Relative Positioning and Dense Reconstruction for UAV Cluster

As robotics technology advances, dense point cloud maps are increasingly in demand. However, dense reconstruction using a single unmanned aerial vehicle (UAV) suffers from limitations in flight speed and battery power, resulting in slow reconstruction and low coverage. Cluster UAV systems offer greater flexibility and wider coverage for map building. Existing methods of cluster UAVs face challenges with accurate relative positioning, scale drift, and high-speed dense point cloud map generation. To address these issues, we propose a cluster framework for large-scale dense reconstruction and real-time collaborative localization. The front-end of the framework is an improved visual odometry which can effectively handle large-scale scenes. Collaborative localization between UAVs is enabled through a two-stage joint optimization algorithm and a relative pose optimization algorithm, effectively achieving accurate relative positioning of UAVs and mitigating scale drift. Estimated poses are used to achieve real-time dense reconstruction and fusion of point cloud maps. To evaluate the performance of our proposed method, we conduct qualitative and quantitative experiments on real-world data. The results demonstrate that our framework can effectively suppress scale drift and generate large-scale dense point cloud maps in real-time, with the reconstruction speed increasing as more UAVs are added to the system.

cs.RO↗

arXiv2023

HybridFusion: LiDAR and Vision Cross-Source Point Cloud Fusion

Recently, cross-source point cloud registration from different sensors has become a significant research focus. However, traditional methods confront challenges due to the varying density and structure of cross-source point clouds. In order to solve these problems, we propose a cross-source point cloud fusion algorithm called HybridFusion. It can register cross-source dense point clouds from different viewing angle in outdoor large scenes. The entire registration process is a coarse-to-fine procedure. First, the point cloud is divided into small patches, and a matching patch set is selected based on global descriptors and spatial distribution, which constitutes the coarse matching process. To achieve fine matching, 2D registration is performed by extracting 2D boundary points from patches, followed by 3D adjustment. Finally, the results of multiple patch pose estimates are clustered and fused to determine the final pose. The proposed approach is evaluated comprehensively through qualitative and quantitative experiments. In order to compare the robustness of cross-source point cloud registration, the proposed method and generalized iterative closest point method are compared. Furthermore, a metric for describing the degree of point cloud filling is proposed. The experimental results demonstrate that our approach achieves state-of-the-art performance in cross-source point cloud registration.

cs.CV↗

arXiv2023

Evolutionary Multi-Objective Virtual Network Function Placement: A Formal Model and Effective Algorithms

Data centers are critical to the commercial and social activities of modern society but are also major electricity consumers. To minimize their environmental impact, it is imperative to make data centers more energy efficient while maintaining a high quality of service (QoS). Bearing this consideration in mind, we develop an analytical model using queueing theory for evaluating the QoS of a data center. Furthermore, based on this model, we develop a domain-specific evolutionary optimization framework featuring a tailored solution representation and a constraint-aware initialization operator for finding the optimal placement of virtual network functions in a data center that optimizes multiple conflicting objectives with regard to energy consumption and QoS. In particular, our framework is applicable to any existing evolutionary multi-objective optimization algorithm in a plug-in manner. Extensive experiments validate the efficiency and accuracy of our QoS model as well as the effectiveness of our tailored algorithms for virtual network function placement problems at various scales.

cs.NI↗

arXiv2023

Multidimensional Resource Fragmentation-Aware Virtual Network Embedding in MEC Systems Interconnected by Metro Optical Networks

The increasing demand for diverse emerging applications has resulted in the interconnection of multi-access edge computing (MEC) systems via metro optical networks. To cater to these diverse applications, network slicing has become a popular tool for creating specialized virtual networks. However, resource fragmentation caused by uneven utilization of multidimensional resources can lead to reduced utilization of limited edge resources. To tackle this issue, this paper focuses on addressing the multidimensional resource fragmentation problem in virtual network embedding (VNE) in MEC systems with the aim of maximizing the profit of an infrastructure provider (InP). The VNE problem in MEC systems is transformed into a bilevel optimization problem, taking into account the interdependence between virtual node embedding (VNoE) and virtual link embedding (VLiE). To solve this problem, we propose a nested bilevel optimization approach named BiVNE. The VNoE is solved using the ant colony system (ACS) in the upper level, while the VLiE is solved using a combination of a shortest path algorithm and an exact-fit spectrum slot allocation method in the lower level. Evaluation results show that the BiVNE algorithm can effectively enhance the profit of the InP by increasing the acceptance ratio and avoiding resource fragmentation simultaneously.

cs.NI↗

arXiv2023

End-To-End Optimization of LiDAR Beam Configuration for 3D Object Detection and Localization

Existing learning methods for LiDAR-based applications use 3D points scanned under a pre-determined beam configuration, e.g., the elevation angles of beams are often evenly distributed. Those fixed configurations are task-agnostic, so simply using them can lead to sub-optimal performance. In this work, we take a new route to learn to optimize the LiDAR beam configuration for a given application. Specifically, we propose a reinforcement learning-based learning-to-optimize (RL-L2O) framework to automatically optimize the beam configuration in an end-to-end manner for different LiDAR-based applications. The optimization is guided by the final performance of the target task and thus our method can be integrated easily with any LiDAR-based application as a simple drop-in module. The method is especially useful when a low-resolution (low-cost) LiDAR is needed, for instance, for system deployment at a massive scale. We use our method to search for the beam configuration of a low-resolution LiDAR for two important tasks: 3D object detection and localization. Experiments show that the proposed RL-L2O method improves the performance in both tasks significantly compared to the baseline methods. We believe that a combination of our method with the recent advances of programmable LiDARs can start a new research direction for LiDAR-based active perception. The code is publicly available at https://github.com/vniclas/lidar_beam_selection

cs.RO↗

arXiv2023

CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentation

Weakly supervised semantic segmentation (WSSS) with image-level labels is a challenging task. Mainstream approaches follow a multi-stage framework and suffer from high training costs. In this paper, we explore the potential of Contrastive Language-Image Pre-training models (CLIP) to localize different categories with only image-level labels and without further training. To efficiently generate high-quality segmentation masks from CLIP, we propose a novel WSSS framework called CLIP-ES. Our framework improves all three stages of WSSS with special designs for CLIP: 1) We introduce the softmax function into GradCAM and exploit the zero-shot ability of CLIP to suppress the confusion caused by non-target classes and backgrounds. Meanwhile, to take full advantage of CLIP, we re-explore text inputs under the WSSS setting and customize two text-driven strategies: sharpness-based prompt selection and synonym fusion. 2) To simplify the stage of CAM refinement, we propose a real-time class-aware attention-based affinity (CAA) module based on the inherent multi-head self-attention (MHSA) in CLIP-ViTs. 3) When training the final segmentation model with the masks generated by CLIP, we introduced a confidence-guided loss (CGL) focus on confident regions. Our CLIP-ES achieves SOTA performance on Pascal VOC 2012 and MS COCO 2014 while only taking 10% time of previous methods for the pseudo mask generation. Code is available at https://github.com/linyq2117/CLIP-ES.

cs.CV↗

arXiv2023

SCADE: NeRFs from Space Carving with Ambiguity-Aware Depth Estimates

Neural radiance fields (NeRFs) have enabled high fidelity 3D reconstruction from multiple 2D input views. However, a well-known drawback of NeRFs is the less-than-ideal performance under a small number of views, due to insufficient constraints enforced by volumetric rendering. To address this issue, we introduce SCADE, a novel technique that improves NeRF reconstruction quality on sparse, unconstrained input views for in-the-wild indoor scenes. To constrain NeRF reconstruction, we leverage geometric priors in the form of per-view depth estimates produced with state-of-the-art monocular depth estimation models, which can generalize across scenes. A key challenge is that monocular depth estimation is an ill-posed problem, with inherent ambiguities. To handle this issue, we propose a new method that learns to predict, for each view, a continuous, multimodal distribution of depth estimates using conditional Implicit Maximum Likelihood Estimation (cIMLE). In order to disambiguate exploiting multiple views, we introduce an original space carving loss that guides the NeRF representation to fuse multiple hypothesized depth maps from each view and distill from them a common geometry that is consistent with all views. Experiments show that our approach enables higher fidelity novel view synthesis from sparse views. Our project page can be found at https://scade-spacecarving-nerfs.github.io .

cs.CV↗

arXiv2023

Practical Cross-System Shilling Attacks with Limited Access to Data

In shilling attacks, an adversarial party injects a few fake user profiles into a Recommender System (RS) so that the target item can be promoted or demoted. Although much effort has been devoted to developing shilling attack methods, we find that existing approaches are still far from practical. In this paper, we analyze the properties a practical shilling attack method should have and propose a new concept of Cross-system Attack. With the idea of Cross-system Attack, we design a Practical Cross-system Shilling Attack (PC-Attack) framework that requires little information about the victim RS model and the target RS data for conducting attacks. PC-Attack is trained to capture graph topology knowledge from public RS data in a self-supervised manner. Then, it is fine-tuned on a small portion of target data that is easy to access to construct fake profiles. Extensive experiments have demonstrated the superiority of PC-Attack over state-of-the-art baselines. Our implementation of PC-Attack is available at https://github.com/KDEGroup/PC-Attack.

cs.IR↗