arXiv subjects

Zihao Zhang

Publications and source records attributed to Zihao Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2023

GICI-LIB: A GNSS/INS/Camera Integrated Navigation Library

Accurate navigation is essential for autonomous robots and vehicles. In recent years, the integration of the Global Navigation Satellite System (GNSS), Inertial Navigation System (INS), and camera has garnered considerable attention due to its robustness and high accuracy in diverse environments. However, leveraging the full capacity of GNSS is cumbersome because of the diverse choices of formulations, error models, satellite constellations, signal frequencies, and service types, which lead to different precision, robustness, and usage dependencies. To clarify the capacity of GNSS algorithms and accelerate the development efficiency of employing GNSS in multi-sensor fusion algorithms, we open source the GNSS/INS/Camera Integration Library (GICI-LIB), together with detailed documentation and a comprehensive land vehicle dataset. A factor graph optimization-based multi-sensor fusion framework is established, which combines almost all GNSS measurement error sources by fully considering temporal and spatial correlations between measurements. The graph structure is designed for flexibility, making it easy to form any kind of integration algorithm. For illustration, Real-Time Kinematic (RTK), Precise Point Positioning (PPP), and four RTK-based algorithms from GICI-LIB are evaluated using our dataset and public datasets. Results confirm the potential of the GICI system to provide continuous precise navigation solutions in a wide spectrum of urban environments.

cs.RO↗

arXiv2023

EHRTutor: Enhancing Patient Understanding of Discharge Instructions

Large language models have shown success as a tutor in education in various fields. Educating patients about their clinical visits plays a pivotal role in patients' adherence to their treatment plans post-discharge. This paper presents EHRTutor, an innovative multi-component framework leveraging the Large Language Model (LLM) for patient education through conversational question-answering. EHRTutor first formulates questions pertaining to the electronic health record discharge instructions. It then educates the patient through conversation by administering each question as a test. Finally, it generates a summary at the end of the conversation. Evaluation results using LLMs and domain experts have shown a clear preference for EHRTutor over the baseline. Moreover, EHRTutor also offers a framework for generating synthetic patient education dialogues that can be used for future in-house system training.

cs.CL↗

arXiv2023

AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism

Generating 3D human motion based on textual descriptions has been a research focus in recent years. It requires the generated motion to be diverse, natural, and conform to the textual description. Due to the complex spatio-temporal nature of human motion and the difficulty in learning the cross-modal relationship between text and motion, text-driven motion generation is still a challenging problem. To address these issues, we propose \textbf{AttT2M}, a two-stage method with multi-perspective attention mechanism: \textbf{body-part attention} and \textbf{global-local motion-text attention}. The former focuses on the motion embedding perspective, which means introducing a body-part spatio-temporal encoder into VQ-VAE to learn a more expressive discrete latent space. The latter is from the cross-modal perspective, which is used to learn the sentence-level and word-level motion-text cross-modal relationship. The text-driven motion is finally generated with a generative transformer. Extensive experiments conducted on HumanML3D and KIT-ML demonstrate that our method outperforms the current state-of-the-art works in terms of qualitative and quantitative evaluation, and achieve fine-grained synthesis and action2motion. Our code is in https://github.com/ZcyMonkey/AttT2M

cs.CV↗

arXiv2023

Subsonic flows with a contact discontinuity in a finitely long axisymmetric cylinder

This paper concerns the structural stability of subsonic flows with a contact discontinuity in a finitely long axisymmetric cylinder. We establish the existence and uniqueness of axisymmetric subsonic flows with a contact discontinuity by prescribing the horizontal mass flux distribution, the swirl velocity, the entropy and the Bernoulli's quantity at the entrance and the radial velocity at the exit. It can be formulated as a free boundary problem with the contact discontinuity to be determined simultaneously with the flows. Compared with the two-dimensional case, a new difficulty arises due to the singularity near the axis. One of the key points in the analysis is the introduction of an invertible modified Lagrangian transformation which can overcome this difficulty and straighten the contact discontinuity. Another one is to utilize the deformation-curl decomposition for the steady Euler system introduced in \cite{WX19} to effectively decouple the hyperbolic and elliptic modes. Finally, the contact discontinuity will be located by using the implicit function theorem.

math.AP↗

arXiv2023

Supersonic flows with a contact discontinuity to the two-dimensional steady rotating Euler system

This paper concerns the structural stability of supersonic flows with a contact discontinuity in a finitely long curved nozzle for the two-dimensional steady compressible rotating Euler system. Concerning the effect of Coriolis force, we first establish the existence of supersonic shear flows with a contact discontinuity in the flat nozzle. Then we consider the stability of these background supersonic shear flows with a contact discontinuity when the incoming supersonic flow and the upper and lower nozzle walls are suitably perturbed. The problem can be formulated as an initial boundary value problem with a contact discontinuity as a free boundary. To deal with the free boundary value problem, the Lagrangian transformation is introduced to straighten and fix the contact discontinuity. The rotating Euler system is reduced to a first order hyperbolic system for the Riemann invariants. We design an iteration scheme and derive some estimates for the solution to the hyperbolic system. Finally, by using the inverse Lagrangian transformation, we prove the original free boundary problem admits two layers of smooth supersonic flows separated by a smooth contact discontinuity.

math.AP↗

arXiv2023

A sufficient condition for $(α, β)$ Somos $4$ Hankel determinants

By using Sulanke-Xin continued fractions method, Xin proposed a recursion system to solve the Somos 4 Hankel determinant conjecture. We find Xin's recursion system indeed give a sufficient condition for $(α, β)$ Somos $4$ sequences. This allows us to prove 4 conjectures of Barry on $(α, β)$ Somos $4$ sequences in a unified way.

math.CO↗

arXiv2023

Pose-aware Attention Network for Flexible Motion Retargeting by Body Part

Motion retargeting is a fundamental problem in computer graphics and computer vision. Existing approaches usually have many strict requirements, such as the source-target skeletons needing to have the same number of joints or share the same topology. To tackle this problem, we note that skeletons with different structure may have some common body parts despite the differences in joint numbers. Following this observation, we propose a novel, flexible motion retargeting framework. The key idea of our method is to regard the body part as the basic retargeting unit rather than directly retargeting the whole body motion. To enhance the spatial modeling capability of the motion encoder, we introduce a pose-aware attention network (PAN) in the motion encoding phase. The PAN is pose-aware since it can dynamically predict the joint weights within each body part based on the input pose, and then construct a shared latent space for each body part by feature pooling. Extensive experiments show that our approach can generate better motion retargeting results both qualitatively and quantitatively than state-of-the-art methods. Moreover, we also show that our framework can generate reasonable results even for a more challenging retargeting scenario, like retargeting between bipedal and quadrupedal skeletons because of the body part retargeting strategy and PAN. Our code is publicly available.

cs.GR↗

arXiv2023

Robots in the Garden: Artificial Intelligence and Adaptive Landscapes

This paper introduces ELUA, the Ecological Laboratory for Urban Agriculture, a collaboration among landscape architects, architects and computer scientists who specialize in artificial intelligence, robotics and computer vision. ELUA has two gantry robots, one indoors and the other outside on the rooftop of a 6-story campus building. Each robot can seed, water, weed, and prune in its garden. To support responsive landscape research, ELUA also includes sensor arrays, an AI-powered camera, and an extensive network infrastructure. This project demonstrates a way to integrate artificial intelligence into an evolving urban ecosystem, and encourages landscape architects to develop an adaptive design framework where design becomes a long-term engagement with the environment.

cs.RO↗

arXiv2023

The Future of Artificial Intelligence (AI) and Machine Learning (ML) in Landscape Design: A Case Study in Coastal Virginia, USA

There have been theory-based endeavours that directly engage with AI and ML in the landscape discipline. By presenting a case that uses machine learning techniques to predict variables in a coastal environment, this paper provides empirical evidence of the forthcoming cybernetic environment, in which designers are conceptualized not as authors but as choreographers, catalyst agents, and conductors among many other intelligent agents. Drawing ideas from posthumanism, this paper argues that, to truly understand the cybernetic environment, we have to take on posthumanist ethics and overcome human exceptionalism.

cs.AI↗

arXiv2023

Cybernetic Environment: A Historical Reflection on System, Design, and Machine Intelligence

Taking on a historical lens, this paper traces the development of cybernetics and systems thinking back to the 1950s, when a group of interdisciplinary scholars converged to create a new theoretical model based on machines and systems for understanding matters of meaning, information, consciousness, and life. By presenting a genealogy of research in the landscape architecture discipline, the paper argues that landscape architects have been an important part of the development of cybernetics by materializing systems based on cybernetic principles in the environment through ecologically based landscape design. The landscape discipline has developed a design framework that provides transformative insights into understanding machine intelligence. The paper calls for a new paradigm of environmental engagement to understand matters of design and machine intelligence.

cs.AI↗

arXiv2023

Cultivated Wildness: Technodiversity and Wildness in Machines

This paper investigates the idea of cultivated wildness at the intersection of landscape design and artificial intelligence. The paper posits that contemporary landscape practices should overcome the potentially single understanding on wilderness, and instead explore landscape strategies to cultivate new forms of wild places via ideas and concerns in contemporary Environmental Humanities, Science and Technology Studies, Ecological Sciences, and Landscape Architecture. Drawing cases in environmental engineering, computer science, and landscape architecture research, this paper explores a framework to construct wild places with intelligent machines. In this framework, machines are not understood as a layer of "digital infrastructure" that is used to extend localized human intelligence and agency. Rather machines are conceptualized as active agents who can participate in the intelligence of co-production. Recent developments in cybernetic technologies such as sensing networks, artificial intelligence, and cyberphysical systems can also contribute to establishing the framework. At the heart of this framework is "technodiversity," in parallel with biodiversity, since a singular vision on technological development driven by optimization and efficiency reinforces a monocultural approach that eliminates other possible relationships to construct with the environment. Thus, cultivated wildness is also about recognizing "wildness" in machines.

cs.AI↗

arXiv2022

Spatio-Temporal Gating-Adjacency GCN for Human Motion Prediction

Predicting future motion based on historical motion sequence is a fundamental problem in computer vision, and it has wide applications in autonomous driving and robotics. Some recent works have shown that Graph Convolutional Networks(GCN) are instrumental in modeling the relationship between different joints. However, considering the variants and diverse action types in human motion data, the cross-dependency of the spatio-temporal relationships will be difficult to depict due to the decoupled modeling strategy, which may also exacerbate the problem of insufficient generalization. Therefore, we propose the Spatio-Temporal Gating-Adjacency GCN(GAGCN) to learn the complex spatio-temporal dependencies over diverse action types. Specifically, we adopt gating networks to enhance the generalization of GCN via the trainable adaptive adjacency matrix obtained by blending the candidate spatio-temporal adjacency matrices. Moreover, GAGCN addresses the cross-dependency of space and time by balancing the weights of spatio-temporal modeling and fusing the decoupled spatio-temporal features. Extensive experiments on Human 3.6M, AMASS, and 3DPW demonstrate that GAGCN achieves state-of-the-art performance in both short-term and long-term predictions.

cs.CV↗

arXiv2022

PNM: Pixel Null Model for General Image Segmentation

A major challenge in image segmentation is classifying object boundaries. Recent efforts propose to refine the segmentation result with boundary masks. However, models are still prone to misclassifying boundary pixels even when they correctly capture the object contours. In such cases, even a perfect boundary map is unhelpful for segmentation refinement. In this paper, we argue that assigning proper prior weights to error-prone pixels such as object boundaries can significantly improve the segmentation quality. Specifically, we present the \textit{pixel null model} (PNM), a prior model that weights each pixel according to its probability of being correctly classified by a random segmenter. Empirical analysis shows that PNM captures the misclassification distribution of different state-of-the-art (SOTA) segmenters. Extensive experiments on semantic, instance, and panoptic segmentation tasks over three datasets (Cityscapes, ADE20K, MS COCO) confirm that PNM consistently improves the segmentation quality of most SOTA methods (including the vision transformers) and outperforms boundary-based methods by a large margin. We also observe that the widely-used mean IoU (mIoU) metric is insensitive to boundaries of different sharpness. As a byproduct, we propose a new metric, \textit{PNM IoU}, which perceives the boundary sharpness and better reflects the model segmentation performance in error-prone regions.

cs.CV↗

arXiv2021

Proof of a Conjecture on Hankel Determinants for Dyck Paths with Restricted Peak Heights

For any integer $m\geq 2$ and $r \in \{1,\dots, m\}$, let $f_n^{m,r}$ denote the number of $n$-Dyck paths whose peak's heights are $im+r$ for some integer $i$. We find the generating function of $f_n^{m,r}$ satisfies a simple algebraic functional equation of degree $2$. The $r=m$ case is particularly nice and we give a combinatorial proof. By using the Sulanke and Xin's continued fraction method, we calculate the Hankel determinants for $f_n^{m,r}$. The special case $r=m$ of our result solves a conjecture proposed by Chien, Eu and Fu. We also enriched the class of eventually periodic Hankel determinant sequences.

math.CO↗

arXiv2021

Deep Open Set Identification for RF Devices

Artificial intelligence (AI) based device identification improves the security of the internet of things (IoT), and accelerates the authentication process. However, existing approaches rely on the assumption that we can learn all the classes from the training set, namely, closed-set classification. To overcome the closed-set limitation, we propose a novel open set RF device identification method to classify unseen classes in the testing set. First, we design a specific convolution neural network (CNN) with a short-time Fourier transforming (STFT) pre-processing module, which efficiently recognizes the differences of feature maps learned from various RF device signals. Then to generate a representation of known class bounds, we estimate the probability map of the open-set via the OpenMax function. We conduct experiments on sampled data and voice signal sets, considering various pre-processing schemes, network structures, distance metrics, tail sizes, and openness degrees. The simulation results show the superiority of the proposed method in terms of robustness and accuracy.

eess.SP↗

arXiv2021

A Universal End-to-End Approach to Portfolio Optimization via Deep Learning

We propose a universal end-to-end framework for portfolio optimization where asset distributions are directly obtained. The designed framework circumvents the traditional forecasting step and avoids the estimation of the covariance matrix, lifting the bottleneck for generalizing to a large amount of instruments. Our framework has the flexibility of optimizing various objective functions including Sharpe ratio, mean-variance trade-off etc. Further, we allow for short selling and study several constraints attached to objective functions. In particular, we consider cardinality, maximum position for individual instrument and leverage. These constraints are formulated into objective functions by utilizing several neural layers and gradient ascent can be adopted for optimization. To ensure the robustness of our framework, we test our methods on two datasets. Firstly, we look at a synthetic dataset where we demonstrate that weights obtained from our end-to-end approach are better than classical predictive methods. Secondly, we apply our framework on a real-life dataset with historical observations of hundreds of instruments with a testing period of more than 20 years.

q-fin.PM↗

arXiv2021

Sweeping Plasma Frequency of Terahertz Surface Plasmon Polaritons with Graphene

Plasma frequency is the spectral boundary for low-loss propagation and evanescent decay of surface plasmon polariton (SPP) waves, which corresponds to a high cut-off phenomenon and is typically utilized for identifying SPPs. At terahertz (THz) frequencies, a metal line with periodic metallic grooves can mimic the conventional optical SPPs, which is referred to as designer SPPs. Theoretically, the plasma frequency of THz SPPs decreases as the groove depth increases. Here, by replacing the metallic grooves with graphene sheets, dynamically sweeping SPP plasma frequency is demonstrated for the first time. The metal-graphene hybrid structure comprises a metal line with periodic graphene grooves, a thin-layer ion gel for gating graphene, and metallic tips for uniforming gate field. As the chemical potential changes, the average conductivity of graphene is modulated so that the effective depth of the graphene grooves changes, which sweeps the plasma frequency of THz SPPs consequently. Both simulated and experimental data demonstrate a red shift of plasma frequency from 195 to 180 GHz at a low bias from -0.5 to 0.5 V. The proposed structure reveals a novel approach to control the on/off status of SPP propagation in the THz range.

physics.optics↗

arXiv2021

Multi-Horizon Forecasting for Limit Order Books: Novel Deep Learning Approaches and Hardware Acceleration using Intelligent Processing Units

We design multi-horizon forecasting models for limit order book (LOB) data by using deep learning techniques. Unlike standard structures where a single prediction is made, we adopt encoder-decoder models with sequence-to-sequence and Attention mechanisms to generate a forecasting path. Our methods achieve comparable performance to state-of-art algorithms at short prediction horizons. Importantly, they outperform when generating predictions over long horizons by leveraging the multi-horizon setup. Given that encoder-decoder models rely on recurrent neural layers, they generally suffer from slow training processes. To remedy this, we experiment with utilising novel hardware, so-called Intelligent Processing Units (IPUs) produced by Graphcore. IPUs are specifically designed for machine intelligence workload with the aim to speed up the computation process. We show that in our setup this leads to significantly faster training times when compared to training models with GPUs.

cs.LG↗