Searcharxiv⌕ Search

arXiv subjects

Hong Wang

Publications and source records attributed to Hong Wang.

At least 199 records · Page 11Linked to original sources

Dimensions of Furstenberg sets and an extension of Bourgain's projection theorem

We show that the Hausdorff dimension of $(s,t)$-Furstenberg sets is at least $s+t/2+ε$, where $ε>0$ depends only on $s$ and $t$. This improves the previously best known bound for $2s<t\le 1+ε(s,t)$, in particular providing the first improvement since 1999 to the dimension of classical $s$-Furstenberg sets for $s<1/2$. We deduce this from a corresponding discretized incidence bound under minimal non-concentration assumptions, that simultaneously extends Bourgain's discretized projection and sum-product theorems. The proofs are based on a recent discretized incidence bound of T.~Orponen and the first author and a certain duality between $(s,t)$ and $(t/2,s+t/2)$-Furstenberg sets.

math.CA↗

SOTIF Entropy: Online SOTIF Risk Quantification and Mitigation for Autonomous Driving

Autonomous driving confronts great challenges in complex traffic scenarios, where the risk of Safety of the Intended Functionality (SOTIF) can be triggered by the dynamic operational environment and system insufficiencies. The SOTIF risk is reflected not only intuitively in the collision risk with objects outside the autonomous vehicles (AVs), but also inherently in the performance limitation risk of the implemented algorithms themselves. How to minimize the SOTIF risk for autonomous driving is currently a critical, difficult, and unresolved issue. Therefore, this paper proposes the "Self-Surveillance and Self-Adaption System" as a systematic approach to online minimize the SOTIF risk, which aims to provide a systematic solution for monitoring, quantification, and mitigation of inherent and external risks. The core of this system is the risk monitoring of the implemented artificial intelligence algorithms within the AV. As a demonstration of the Self-Surveillance and Self-Adaption System, the risk monitoring of the perception algorithm, i.e., YOLOv5 is highlighted. Moreover, the inherent perception algorithm risk and external collision risk are jointly quantified via SOTIF entropy, which is then propagated downstream to the decision-making module and mitigated. Finally, several challenging scenarios are demonstrated, and the Hardware-in-the-Loop experiments are conducted to verify the efficiency and effectiveness of the system. The results demonstrate that the Self-Surveillance and Self-Adaption System enables dependable online monitoring, quantification, and mitigation of SOTIF risk in real-time critical traffic environments.

cs.AI↗

PeSOTIF: a Challenging Visual Dataset for Perception SOTIF Problems in Long-tail Traffic Scenarios

Perception algorithms in autonomous driving systems confront great challenges in long-tail traffic scenarios, where the problems of Safety of the Intended Functionality (SOTIF) could be triggered by the algorithm performance insufficiencies and dynamic operational environment. However, such scenarios are not systematically included in current open-source datasets, and this paper fills the gap accordingly. Based on the analysis and enumeration of trigger conditions, a high-quality diverse dataset is released, including various long-tail traffic scenarios collected from multiple resources. Considering the development of probabilistic object detection (POD), this dataset marks trigger sources that may cause perception SOTIF problems in the scenarios as key objects. In addition, an evaluation protocol is suggested to verify the effectiveness of POD algorithms in identifying the key objects via uncertainty. The dataset never stops expanding, and the first batch of open-source data includes 1126 frames with an average of 2.27 key objects and 2.47 normal objects in each frame. To demonstrate how to use this dataset for SOTIF research, this paper further quantifies the perception SOTIF entropy to confirm whether a scenario is unknown and unsafe for a perception system. The experimental results show that the quantified entropy can effectively and efficiently reflect the failure of the perception algorithm.

cs.CV↗

Explanations from Large Language Models Make Small Reasoners Better

Integrating free-text explanations to in-context learning of large language models (LLM) is shown to elicit strong reasoning capabilities along with reasonable explanations. In this paper, we consider the problem of leveraging the explanations generated by LLM to improve the training of small reasoners, which are more favorable in real-production deployment due to their low cost. We systematically explore three explanation generation approaches from LLM and utilize a multi-task learning framework to facilitate small models to acquire strong reasoning power together with explanation generation capabilities. Experiments on multiple reasoning tasks show that our method can consistently and significantly outperform finetuning baselines across different settings, and even perform better than finetuning/prompting a 60x larger GPT-3 (175B) model by up to 9.5% in accuracy. As a side benefit, human evaluation further shows that our method can generate high-quality explanations to justify its predictions, moving towards the goal of explainable AI.

cs.CL↗

Towards Theoretically Inspired Neural Initialization Optimization

Automated machine learning has been widely explored to reduce human efforts in designing neural architectures and looking for proper hyperparameters. In the domain of neural initialization, however, similar automated techniques have rarely been studied. Most existing initialization methods are handcrafted and highly dependent on specific architectures. In this paper, we propose a differentiable quantity, named GradCosine, with theoretical insights to evaluate the initial state of a neural network. Specifically, GradCosine is the cosine similarity of sample-wise gradients with respect to the initialized parameters. By analyzing the sample-wise optimization landscape, we show that both the training and test performance of a network can be improved by maximizing GradCosine under gradient norm constraint. Based on this observation, we further propose the neural initialization optimization (NIO) algorithm. Generalized from the sample-wise analysis into the real batch setting, NIO is able to automatically look for a better initialization with negligible cost compared with the training time. With NIO, we improve the classification performance of a variety of neural architectures on CIFAR-10, CIFAR-100, and ImageNet. Moreover, we find that our method can even help to train large vision Transformer architecture without warmup.

cs.LG↗

Linear Optical Modulators for Prospective Communications at the 2 μm Waveband

The 2 μm waveband is an area that could have significant technological consequences, with applications ranging from spectroscopy, LIDAR and free-space communications. The development of the thulium-doped fiber amplifier, hollow-core photonic bandgap fiber and 2 μm GaSb-based diode lasers has highlighted the ability of the waveband in alleviating the fiber capacity crisis in the incumbent communication infrastructure. The above has initiated vibrant development in the silicon photonic-space at 2 μm, where the area is capable of enabling highly-integrated photonic circuits, and potentially at low-cost and high-volumes. However, as of now, modulator linearity at 2 μm has not been addressed. The metric, as characterized by spurious free dynamic range is imperative for numerous applications such as RF photonic links for 5G and digital analog transmission in coherent communications. The development of linear optical modulators will be crucial in bringing these applications to the 2 μm. In view of that, this work is the first to address modulator linearity at the 2 μm, where the ring-assisted Mach-Zehnder modulator is developed, indicating spurious free dynamic range as high as 95 dB.Hz^2/3. It is found that that modulator spurious free dynamic range has a strong dependence on modulator bias voltage where it impacts the linearity of the transfer function in which the input RF signal is applied upon. The demonstrated modulator indicates favorable performance within silicon photonic modulators developed at 2 μm with bandwidth exceeding 17.5 GHz and modulation efficiency ranging from 0.70 to 1.25 V.cm.

physics.app-ph↗

The facet ideals of chessboard complexes

In this paper we describe the irreducible decomposition of the facet ideal $\F(Δ_{m,n})$ of the chessboard complex $Δ_{m,n}$ with $n\geq m$. We also provide some lower bounds for depth and regularity of the facet ideal $\F(Δ_{m,n})$. When $m\leq 3$, we prove that these lower bounds can be obtained.

math.AC↗

Performance Analysis of MIMO-NOMA Systems with Randomly Deployed Users

This paper investigates the performance of Multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) systems with randomly deployed users, where the randomly deployed NOMA users follow Poisson point process (PPP), the spatial correlation between MIMO channels are characterized by using Kronecker model, and the composite channel model is used to capture large-scale fading as well as small-scale fading. The spatial randomness of users' distribution, the spatial correlation among antennas and large-scale fading will severally impact the system performance, but they are seldom considered in prior literature for MIMO-NOMA systems, and the consideration of all these impact factors challenges the analysis. Based on zero-forcing (ZF) detection, the exact expressions for both the average outage probability and the average goodput are derived in closed-form. Moreover, the asymptotic analyses are conducted for both high signal-to-noise ratio (SNR) (/small cell radius) and low SNR (/large cell radius) to gain more insightful results. In particular, the diversity order is given by $δ= {{N_r} - {M} + 1}$, the average outage probability of $k$-th nearest user to the base station follows a scaling law of $O\left({D^{α\left( {{N_r} - {M} + 1} \right) + 2k}}\right)$, the average goodput scales as $O({D^{2}})$ and $O({D^{-2}})$ as $D \to 0$ and $D \to \infty$, respectively, where $N_r$, $M$, $α$ and $D$ stand for the number of receive antennas, the total number of data streams, the path loss exponent and the cell radius, respectively. The analytical results are finally validated through the numerical analysis.

cs.IT↗

Zero-Forcing Based Downlink Virtual MIMO-NOMA Communications in IoT Networks

To support massive connectivity and boost spectral efficiency for internet of things (IoT), a downlink scheme combining virtual multiple-input multiple-output (MIMO) and nonorthogonal multiple access (NOMA) is proposed. All the single-antenna IoT devices in each cluster cooperate with each other to establish a virtual MIMO entity, and multiple independent data streams are requested by each cluster. NOMA is employed to superimpose all the requested data streams, and each cluster leverages zero-forcing detection to de-multiplex the input data streams. Only statistical channel state information (CSI) is available at base station to avoid the waste of the energy and bandwidth on frequent CSI estimations. The outage probability and goodput of the virtual MIMO-NOMA system are thoroughly investigated by considering Kronecker model, which embraces both the transmit and receive correlations. Furthermore, the asymptotic results facilitate not only the exploration of physical insights but also the goodput maximization. In particular, the asymptotic outage expressions provide quantitative impacts of various system parameters and enable the investigation of diversity-multiplexing tradeoff (DMT). Moreover, power allocation coefficients and/or transmission rates can be properly chosen to achieve the maximal goodput. By favor of Karush-Kuhn-Tucker conditions, the goodput maximization problems can be solved in closed-form, with which the joint power and rate selection is realized by using alternately iterating optimization.Besides, the optimization algorithms tend to allocate more power to clusters under unfavorable channel conditions and support clusters with higher transmission rate under benign channel conditions.

cs.IT↗

Achievable Diversity Order of HARQ-Aided Downlink NOMA Systems

The combination between non-orthogonal multiple access (NOMA) and hybrid automatic repeat request (HARQ) is capable of realizing ultra-reliability, high throughput and many concurrent connections particularly for emerging communication systems. This paper focuses on characterizing the asymptotic scaling law of the outage probability of HARQ-aided NOMA systems with respect to the transmit power, i.e., diversity order. The analysis of diversity order is carried out for three basic types of HARQ-aided downlink NOMA systems, including Type I HARQ, HARQ with chase combining (HARQ-CC) and HARQ with incremental redundancy (HARQ-IR). The diversity orders of three HARQ-aided downlink NOMA systems are derived in closed-form, where an integration domain partition trick is developed to obtain the bounds of the outage probability specially for HARQ-CC and HARQ-IR-aided NOMA systems. The analytical results show that the diversity order is a decreasing step function of transmission rate, and full time diversity can only be achieved under a sufficiently low transmission rate. It is also revealed that HARQ-IR-aided NOMA systems have the largest diversity order, followed by HARQ-CC-aided and then Type I HARQ-aided NOMA systems. Additionally, the users' diversity orders follow a descending order according to their respective average channel gains. Furthermore, we expand discussions on the cases of power-efficient transmissions and imperfect channel state information (CSI). Monte Carlo simulations finally confirm our analysis.

cs.IT↗

Outage Performance and Optimal Design of MIMO-NOMA Enhanced Small Cell Networks With Imperfect Channel-State Information

This paper focuses on boosting the performance of small cell networks (SCNs) by integrating multiple-input multiple-output (MIMO) and non-orthogonal multiple access (NOMA) in consideration of imperfect channel-state information (CSI). The estimation error and the spatial randomness of base stations (BSs) are characterized by using Kronecker model and Poisson point process (PPP), respectively. The outage probabilities of MIMO-NOMA enhanced SCNs are first derived in closed-form by taking into account two grouping policies, including random grouping and distance-based grouping. It is revealed that the average outage probabilities are irrelevant to the intensity of BSs in the interference-limited regime, while the outage performance deteriorates if the intensity is sufficiently low. Besides, as the channel uncertainty lessens, the asymptotic analyses manifest that the target rates must be restricted up to a bound to achieve an arbitrarily low outage probability in the absence of the inter-cell interference.Moreover, highly correlated estimation error ameliorates the outage performance under a low quality of CSI, otherwise it behaves oppositely. Afterwards, the goodput is maximized by choosing appropriate precoding matrix, receiver filters and transmission rates. In the end, the numerical results verify our analysis and corroborate the superiority of our proposed algorithm.

cs.IT↗

Outage Performance Analysis of HARQ-Aided Multi-RIS Systems

Reconfigurable intelligent surface (RIS) has recently attracted a spurt of interest due to its innate advantages over Massive MIMO on power consumption. In this paper, we study the outage performance of multi-RIS system with the help of hybrid automatic repeat request (HARQ) to improve the RIS system reliability, where the destination received channels are modeled by Rician fading and the phase shift setting only depends on the line-of-sight (LoS) component. Both the exact and asymptotic outage probabilities under Type-I HARQ and HARQ with chase combining (HARQ-CC) schemes are derived. Particulary, the tractable asymptotic results empower us to derive meaningful insights for HARQ-aided multi-RIS system. On the one hand, we find that both the Type-I and the HARQ-CC schemes can achieve full diversity that is equal to the maximal number of HARQ rounds. On the other hand, the closed-form expression of the optimal phase shift setting with respect to outage probability minimization is obtained. The optimal solution indicates that the reflecting link direction should be consistent with direct link LoS component. Finally, the analytical results are validated by Monte-Carlo simulations.

cs.IT↗

Outage Performance Analysis of Full-Correlated Rayleigh MIMO Channels

The outage performance of multiple-input multiple-output (MIMO) technique has received intensive attention to meet the stringent requirement of reliable communications for 5G applications, e.g., mission-critical machine-type communication (cMTC). To account for spatial correlation effects at both transmit and receive sides, the full-correlated Rayleigh MIMO fading channels are modeled according to Kronecker correlation structure in this paper. The outage probability is expressed as a weighted sum of the generalized Fox's H functions. The simple analytical result empowers asymptotic outage analysis at high signal-to-noise ratio (SNR), which not only reveal helpful insights into understanding the behavior of fading effects, but also offer useful design guideline for MIMO configurations. Particularly, the negative impact of the spatial correlation on the outage probability is revealed by using the concept of majorization, and the asymptotic outage probability is proved to be a monotonically increasing and convex function of the transmission rate. In the end, the analytical results are validated through extensive numerical experiments.

cs.IT↗

KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution

Although current deep learning-based methods have gained promising performance in the blind single image super-resolution (SISR) task, most of them mainly focus on heuristically constructing diverse network architectures and put less emphasis on the explicit embedding of the physical generation mechanism between blur kernels and high-resolution (HR) images. To alleviate this issue, we propose a model-driven deep neural network, called KXNet, for blind SISR. Specifically, to solve the classical SISR model, we propose a simple-yet-effective iterative algorithm. Then by unfolding the involved iterative steps into the corresponding network module, we naturally construct the KXNet. The main specificity of the proposed KXNet is that the entire learning process is fully and explicitly integrated with the inherent physical mechanism underlying this SISR task. Thus, the learned blur kernel has clear physical patterns and the mutually iterative process between blur kernel and HR image can soundly guide the KXNet to be evolved in the right direction. Extensive experiments on synthetic and real data finely demonstrate the superior accuracy and generality of our method beyond the current representative state-of-the-art blind SISR methods. Code is available at: https://github.com/jiahong-fu/KXNet.

cs.CV↗

CAMO-MOT: Combined Appearance-Motion Optimization for 3D Multi-Object Tracking with Camera-LiDAR Fusion

3D Multi-object tracking (MOT) ensures consistency during continuous dynamic detection, conducive to subsequent motion planning and navigation tasks in autonomous driving. However, camera-based methods suffer in the case of occlusions and it can be challenging to accurately track the irregular motion of objects for LiDAR-based methods. Some fusion methods work well but do not consider the untrustworthy issue of appearance features under occlusion. At the same time, the false detection problem also significantly affects tracking. As such, we propose a novel camera-LiDAR fusion 3D MOT framework based on the Combined Appearance-Motion Optimization (CAMO-MOT), which uses both camera and LiDAR data and significantly reduces tracking failures caused by occlusion and false detection. For occlusion problems, we are the first to propose an occlusion head to select the best object appearance features multiple times effectively, reducing the influence of occlusions. To decrease the impact of false detection in tracking, we design a motion cost matrix based on confidence scores which improve the positioning and object prediction accuracy in 3D space. As existing multi-object tracking methods only consider a single category, we also propose to build a multi-category loss to implement multi-object tracking in multi-category scenes. A series of validation experiments are conducted on the KITTI and nuScenes tracking benchmarks. Our proposed method achieves state-of-the-art performance and the lowest identity switches (IDS) value (23 for Car and 137 for Pedestrian) among all multi-modal MOT methods on the KITTI test dataset. And our proposed method achieves state-of-the-art performance among all algorithms on the nuScenes test dataset with 75.3% AMOTA.

cs.CV↗

SIND: A Drone Dataset at Signalized Intersection in China

Intersection is one of the most challenging scenarios for autonomous driving tasks. Due to the complexity and stochasticity, essential applications (e.g., behavior modeling, motion prediction, safety validation, etc.) at intersections rely heavily on data-driven techniques. Thus, there is an intense demand for trajectory datasets of traffic participants (TPs) in intersections. Currently, most intersections in urban areas are equipped with traffic lights. However, there is not yet a large-scale, high-quality, publicly available trajectory dataset for signalized intersections. Therefore, in this paper, a typical two-phase signalized intersection is selected in Tianjin, China. Besides, a pipeline is designed to construct a Signalized INtersection Dataset (SIND), which contains 7 hours of recording including over 13,000 TPs with 7 types. Then, the behaviors of traffic light violations in SIND are recorded. Furthermore, the SIND is also compared with other similar works. The features of the SIND can be summarized as follows: 1) SIND provides more comprehensive information, including traffic light states, motion parameters, High Definition (HD) map, etc. 2) The category of TPs is diverse and characteristic, where the proportion of vulnerable road users (VRUs) is up to 62.6% 3) Multiple traffic light violations of non-motor vehicles are shown. We believe that SIND would be an effective supplement to existing datasets and can promote related research on autonomous driving.The dataset is available online via: https://github.com/SOTIF-AVLab/SinD

cs.CV↗

Kaufman and Falconer estimates for radial projections and a continuum version of Beck's Theorem

We provide several new answers on the question: how do radial projections distort the dimension of planar sets? Let $X,Y \subset \mathbb{R}^{2}$ be non-empty Borel sets. If $X$ is not contained on any line, we prove that \[ \sup_{x \in X} \dim_{\mathrm{H}} π_{x}(Y) \geq \min\{\dim_{\mathrm{H}} X,\dim_{\mathrm{H}} Y,1\}. \] If $\dim_{\mathrm{H}} Y > 1$, we have the following improved lower bound: \[ \sup_{x \in X} \dim_{\mathrm{H}} π_{x}(Y \, \setminus \, \{x\}) \geq \min\{\dim_{\mathrm{H}} X + \dim_{\mathrm{H}} Y - 1,1\}. \] Our results solve conjectures of Lund-Thang-Huong, Liu, and the first author. Another corollary is the following continuum version of Beck's theorem in combinatorial geometry: if $X \subset \mathbb{R}^{2}$ is a Borel set with the property that $\dim_{\mathrm{H}} (X \, \setminus \, \ell) = \dim_{\mathrm{H}} X$ for all lines $\ell \subset \mathbb{R}^{2}$, then the line set spanned by $X$ has Hausdorff dimension at least $\min\{2\dim_{\mathrm{H}} X,2\}$. While the results above concern $\mathbb{R}^{2}$, we also derive some counterparts in $\mathbb{R}^{d}$ by means of integralgeometric considerations. The proofs are based on an $ε$-improvement in the Furstenberg set problem, due to the two first authors, a bootstrapping scheme introduced by the second and third author, and a new planar incidence estimate due to Fu and Ren.

math.CA↗