arXiv subjects

Liang Yu

Publications and source records attributed to Liang Yu.

At least 37 records · Page 2Linked to original sources

arXiv2025

A Low-cost and Ultra-lightweight Binary Neural Network for Traffic Signal Recognition

The deployment of neural networks in vehicle platforms and wearable Artificial Intelligence-of-Things (AIOT) scenarios has become a research area that has attracted much attention. With the continuous evolution of deep learning technology, many image classification models are committed to improving recognition accuracy, but this is often accompanied by problems such as large model resource usage, complex structure, and high power consumption, which makes it challenging to deploy on resource-constrained platforms. Herein, we propose an ultra-lightweight binary neural network (BNN) model designed for hardware deployment, and conduct image classification research based on the German Traffic Sign Recognition Benchmark (GTSRB) dataset. In addition, we also verify it on the Chinese Traffic Sign (CTS) and Belgian Traffic Sign (BTS) datasets. The proposed model shows excellent recognition performance with an accuracy of up to 97.64%, making it one of the best performing BNN models in the GTSRB dataset. Compared with the full-precision model, the accuracy loss is controlled within 1%, and the parameter storage overhead of the model is only 10% of that of the full-precision model. More importantly, our network model only relies on logical operations and low-bit width fixed-point addition and subtraction operations during the inference phase, which greatly simplifies the design complexity of the processing element (PE). Our research shows the great potential of BNN in the hardware deployment of computer vision models, especially in the field of computer vision tasks related to autonomous driving.

cs.AI↗

arXiv2024

Model Based and Physics Informed Deep Learning Neural Network Structures

Neural Networks (NN) has been used in many areas with great success. When a NN's structure (Model) is given, during the training steps, the parameters of the model are determined using an appropriate criterion and an optimization algorithm (Training). Then, the trained model can be used for the prediction or inference step (Testing). As there are also many hyperparameters, related to the optimization criteria and optimization algorithms, a validation step is necessary before its final use. One of the great difficulties is the choice of the NN's structure. Even if there are many "on the shelf" networks, selecting or proposing a new appropriate network for a given data, signal or image processing, is still an open problem. In this work, we consider this problem using model based signal and image processing and inverse problems methods. We classify the methods in five classes, based on: i) Explicit analytical solutions, ii) Transform domain decomposition, iii) Operator Decomposition, iv) Optimization algorithms unfolding, and v) Physics Informed NN methods (PINN). Few examples in each category are explained.

cs.LG↗

arXiv2023

The Point-to-Set Principle and the Dimensions of Hamel Bases

We prove that every real number in [0,1] is the Hausdorff dimension of a Hamel basis of the vector space of reals over the field of rationals. The logic of our proof is of particular interest. The statement of our theorem is classical; it does not involve the theory of computing. However, our proof makes essential use of algorithmic fractal dimension--a computability-theoretic construct--and the point-to-set principle of J. Lutz and N. Lutz (2018).

cs.LO↗

arXiv2023

Computational Approaches for Predicting Drug-Disease Associations: A Comprehensive Review

In recent decades, traditional drug research and development have been facing challenges such as high cost, long timelines, and high risks. To address these issues, many computational approaches have been suggested for predicting the relationship between drugs and diseases through drug repositioning, aiming to reduce the cost, development cycle, and risks associated with developing new drugs. Researchers have explored different computational methods to predict drug-disease associations, including drug side effects-disease associations, drug-target associations, and miRNAdisease associations. In this comprehensive review, we focus on recent advances in predicting drug-disease association methods for drug repositioning. We first categorize these methods into several groups, including neural network-based algorithms, matrixbased algorithms, recommendation algorithms, link-based reasoning algorithms, and text mining and semantic reasoning. Then, we compare the prediction performance of existing drug-disease association prediction algorithms. Lastly, we delve into the present challenges and future prospects concerning drug-disease associations.

q-bio.QM↗

arXiv2023

Deep Learning and Inverse Problems

Machine Learning (ML) methods and tools have gained great success in many data, signal, image and video processing tasks, such as classification, clustering, object detection, semantic segmentation, language processing, Human-Machine interface, etc. In computer vision, image and video processing, these methods are mainly based on Neural Networks (NN) and in particular Convolutional NN (CNN), and more generally Deep NN. Inverse problems arise anywhere we have indirect measurement. As, in general, those inverse problems are ill-posed, to obtain satisfactory solutions for them needs prior information. Different regularization methods have been proposed, where the problem becomes the optimization of a criterion with a likelihood term and a regularization term. The main difficulty, however, in great dimensional real applications, remains the computational cost. Using NN, and in particular Deep Learning (DL) surrogate models and approximate computation, can become very helpful. In this work, we focus on NN and DL particularly adapted for inverse problems. We consider two cases: First the case where the forward operator is known and used as physics constraint, the second more general data driven DL methods.

cs.LG↗

arXiv2023

Some more results on relativized Chaitin's $Ω$

We prove that, assuming $\mathrm{ZF}$, and restricted to any pointed set, Chaitin's $Ω_U:x\mapsto Ω_U^x=\sum_{U^x(σ)\downarrow}2^{-|σ|}$ is not injective for any universal prefix-free Turing machine $U$, and that $Ω_U^x$ fails to be degree invariant in a very strong sense, answering several recent questions in descriptive set theory. Moreover, we show that under $\mathrm{ZF}+\mathrm{AD}$, every function $f$ mapping $x$ to $x$-random must be uncountable-to-one over an upper cone of Turing degrees.

math.LO↗

arXiv2023

Deep Learning and Bayesian inference for Inverse Problems

Inverse problems arise anywhere we have indirect measurement. As, in general they are ill-posed, to obtain satisfactory solutions for them needs prior knowledge. Classically, different regularization methods and Bayesian inference based methods have been proposed. As these methods need a great number of forward and backward computations, they become costly in computation, in particular, when the forward or generative models are complex and the evaluation of the likelihood becomes very costly. Using Deep Neural Network surrogate models and approximate computation can become very helpful. However, accounting for the uncertainties, we need first understand the Bayesian Deep Learning and then, we can see how we can use them for inverse problems. In this work, we focus on NN, DL and more specifically the Bayesian DL particularly adapted for inverse problems. We first give details of Bayesian DL approximate computations with exponential families, then we will see how we can use them for inverse problems. We consider two cases: First the case where the forward operator is known and used as physics constraint, the second more general data driven DL methods. keyword: Neural Network, Variational Bayesian inference, Bayesian Deep Learning (DL), Inverse problems, Physics based DL.

stat.ML↗

arXiv2023

BARS: A Benchmark for Airport Runway Segmentation

Airport runway segmentation can effectively reduce the accident rate during the landing phase, which has the largest risk of flight accidents. With the rapid development of deep learning (DL), related methods achieve good performance on segmentation tasks and can be well adapted to complex scenes. However, the lack of large-scale, publicly available datasets in this field makes the development of methods based on DL difficult. Therefore, we propose a benchmark for airport runway segmentation, named BARS. Additionally, a semiautomatic annotation pipeline is designed to reduce the annotation workload. BARS has the largest dataset with the richest categories and the only instance annotation in the field. The dataset, which was collected using the X-Plane simulation platform, contains 10,256 images and 30,201 instances with three categories. We evaluate eleven representative instance segmentation methods on BARS and analyze their performance. Based on the characteristic of an airport runway with a regular shape, we propose a plug-and-play smoothing postprocessing module (SPM) and a contour point constraint loss (CPCL) function to smooth segmentation results for mask-based and contour-based methods, respectively. Furthermore, a novel evaluation metric named average smoothness (AS) is developed to measure smoothness. The experiments show that existing instance segmentation methods can achieve prediction results with good performance on BARS. SPM and CPCL can effectively enhance the AS metric while modestly improving accuracy. Our work will be available at https://github.com/c-wenhui/BARS.

cs.CV↗

arXiv2023

Road Extraction with Satellite Images and Partial Road Maps

Road extraction is a process of automatically generating road maps mainly from satellite images. Existing models all target to generate roads from the scratch despite that a large quantity of road maps, though incomplete, are publicly available (e.g. those from OpenStreetMap) and can help with road extraction. In this paper, we propose to conduct road extraction based on satellite images and partial road maps, which is new. We then propose a two-branch Partial to Complete Network (P2CNet) for the task, which has two prominent components: Gated Self-Attention Module (GSAM) and Missing Part (MP) loss. GSAM leverages a channel-wise self-attention module and a gate module to capture long-range semantics, filter out useless information, and better fuse the features from two branches. MP loss is derived from the partial road maps, trying to give more attention to the road pixels that do not exist in partial road maps. Extensive experiments are conducted to demonstrate the effectiveness of our model, e.g. P2CNet achieves state-of-the-art performance with the IoU scores of 70.71% and 75.52%, respectively, on the SpaceNet and OSM datasets.

cs.CV↗

arXiv2023

Dynamic Offloading Loading Optimization in distributed Fault Diagnosis system with Deep Reinforcement Learning Approach

Artificial intelligence and distributed algorithms have been widely used in mechanical fault diagnosis with the explosive growth of diagnostic data. A novel intelligent fault diagnosis system framework that allows intelligent terminals to offload computational tasks to Mobile edge computing (MEC) servers is provided in this paper, which can effectively address the problems of task processing delays and enhanced computational complexity. As the resources at the MEC and intelligent terminals are limited, performing reasonable resource allocation optimization can improve the performance, especially for a multi-terminals offloading system. In this study, to minimize the task computation delay, we jointly optimize the local content splitting ratio, the transmission/computation power allocation, and the MEC server selection under a dynamic environment with stochastic task arrivals. The challenging dynamic joint optimization problem is formulated as a reinforcement learning (RL) problem, which is designed as the computational offloading policies to minimize the long-term average delay cost. Two deep RL strategies, deep Q-learning network (DQN) and deep deterministic policy gradient (DDPG), are adopted to learn the computational offloading policies adaptively and efficiently. The proposed DQN strategy takes the MEC selection as a unique action while using the convex optimization approach to obtain the local content splitting ratio and the transmission/computation power allocation. Simultaneously, the actions of the DDPG strategy are selected as all dynamic variables, including the local content splitting ratio, the transmission/computation power allocation, and the MEC server selection. Numerical results demonstrate that both proposed strategies perform better than the traditional non-learning schemes.

cs.IT↗

arXiv2022

A Survey of ADMM Variants for Distributed Optimization: Problems, Algorithms and Features

By coordinating terminal smart devices or microprocessors to engage in cooperative computation to achieve systemlevel targets, distributed optimization is incrementally favored by both engineering and computer science. The well-known alternating direction method of multipliers (ADMM) has turned out to be one of the most popular tools for distributed optimization due to many advantages, such as modular structure, superior convergence, easy implementation and high flexibility. In the past decade, ADMM has experienced widespread developments. The developments manifest in both handling more general problems and enabling more effective implementation. Specifically, the method has been generalized to broad classes of problems (i.e.,multi-block, coupled objective, nonconvex, etc.). Besides, it has been extensively reinforced for more effective implementation, such as improved convergence rate, easier subproblems, higher computation efficiency, flexible communication, compatible with inaccurate information, robust to communication delays, etc. These developments lead to a plentiful of ADMM variants to be celebrated by broad areas ranging from smart grids, smart buildings, wireless communications, machine learning and beyond. However, there lacks a survey to document those developments and discern the results. To achieve such a goal, this paper provides a comprehensive survey on ADMM variants. Particularly, we discern the five major classes of problems that have been mostly concerned and discuss the related ADMM variants in terms of main ideas, main assumptions, convergence behaviors and main features. In addition, we figure out several important future research directions to be addressed. This survey is expected to work as a tutorial for both developing distributed optimization in broad areas and identifying existing theoretical research gaps.

cs.DC↗

arXiv2021

The TESS Faint Star Search: 1,617 TOIs from the TESS Primary Mission

We present the detection of 1,617 new transiting planet candidates, identified in the Transiting Exoplanet Survey Satellite (TESS) full-frame images (FFIs) observed during the Primary Mission (Sectors 1 - 26). These candidates were initially detected by the Quick-Look Pipeline (QLP), which extracts FFI lightcurves for and searches all stars brighter than TESS magnitude T = 13.5 mag in each sector. However, QLP heavily relies on manual inspection for the identification of planet candidates, limiting vetting efforts to planet-hosting stars brighter than T = 10.5 mag and leaving millions of potential transit signals un-vetted. We describe an independent vetting pipeline applied to QLP transit search results, incorporating both automated vetting tests and manual inspection to identify promising planet candidates around these fainter stars. The new candidates discovered by this ongoing project will allow TESS to significantly improve the statistical power of demographics studies of giant, close-in exoplanets.

astro-ph.EP↗

arXiv2021

Optimal Operation of a Hydrogen-based Building Multi-Energy System Based on Deep Reinforcement Learning

Since hydrogen has many advantages (e.g., free pollution, extensive sources, convenient storage and transportation), hydrogen-based multi-energy systems (HMESs) have received wide attention. However, existing works on the optimal operation of HMESs neglect building thermal dynamics, which means that the flexibility of building thermal loads can not be utilized for reducing system operation cost. In this paper, we investigate an optimal operation problem of an HMES with the consideration of building thermal dynamics. Specifically, we first formulate an expected operational cost minimization problem related to an HMES. Due to the existence of uncertain parameters, inexplicit building thermal dynamics models, temporally coupled operational constraints related to three kinds of energy storage systems and indoor temperatures, as well as the coupling between electric energy subsystems and thermal energy subsystems, it is challenging to solve the formulated problem. To overcome the challenge, we reformulate the problem as a Markov game and propose an energy management algorithm to solve it based on multi-agent discrete actor-critic with rules (MADACR). Note that the proposed algorithm does not require any prior knowledge of uncertain parameters, parameter prediction, and explicit building thermal dynamics model. Simulation results based on real-world traces show the effectiveness of the proposed algorithm.

eess.SY↗

arXiv2021

A Review of Deep Reinforcement Learning for Smart Building Energy Management

Global buildings account for about 30% of the total energy consumption and carbon emission, raising severe energy and environmental concerns. Therefore, it is significant and urgent to develop novel smart building energy management (SBEM) technologies for the advance of energy-efficient and green buildings. However, it is a nontrivial task due to the following challenges. Firstly, it is generally difficult to develop an explicit building thermal dynamics model that is both accurate and efficient enough for building control. Secondly, there are many uncertain system parameters (e.g., renewable generation output, outdoor temperature, and the number of occupants). Thirdly, there are many spatially and temporally coupled operational constraints. Fourthly, building energy optimization problems can not be solved in real-time by traditional methods when they have extremely large solution spaces. Fifthly, traditional building energy management methods have respective applicable premises, which means that they have low versatility when confronted with varying building environments. With the rapid development of Internet of Things technology and computation capability, artificial intelligence technology find its significant competence in control and optimization. As a general artificial intelligence technology, deep reinforcement learning (DRL) is promising to address the above challenges. Notably, the recent years have seen the surge of DRL for SBEM. However, there lacks a systematic overview of different DRL methods for SBEM. To fill the gap, this paper provides a comprehensive review of DRL for SBEM from the perspective of system scale. In particular, we identify the existing unresolved issues and point out possible future research directions.

eess.SY↗

arXiv2021

The TESS Mission Target Selection Procedure

We describe the target selection procedure by which stars are selected for 2-minute and 20-second observations by TESS. We first list the technical requirements of the TESS instrument and ground systems processing that limit the total number of target slots. We then describe algorithms used by the TESS Payload Operation Center (POC) to merge candidate targets requested by the various TESS mission elements (the Target Selection Working Group, TESS Asteroseismic Science Consortium, and Guest Investigator office). Lastly, we summarize the properties of the observed TESS targets over the two-year primary TESS mission. We find that the POC target selection algorithm results in 2.1 to 3.4 times as many observed targets as target slots allocated for each mission element. We also find that the sky distribution of observed targets is different from the sky distributions of candidate targets due to technical constraints that require a relatively even distribution of targets across the TESS fields of view. We caution researchers exploring statistical analyses of TESS planet-host stars that the population of observed targets cannot be characterized by any simple set of criteria applied to the properties of the input Candidate Target Lists.

astro-ph.IM↗

arXiv2021

Some consequences of $\mathrm{TD}$ and $\mathrm{sTD}$

Strongly Turing determinacy, or $\mathrm{sTD}$, says that for any set $A$ of reals, if $\forall x\exists y\geq_T x (y\in A)$, then there is a pointed set $P\subseteq A$. We prove the following consequences of Turing determinacy ($\mathrm{TD}$) and $\mathrm{sTD}$: (1). $\mathrm{ZF+TD}$ implies weakly dependent choice ($\mathrm{wDC}$). (2). $\mathrm{ZF+sTD}$ implies that every set of reals is measurable and has Baire property. (3). $\mathrm{ZF+sTD}$ implies that every uncountable set of reals has a perfect subset. (4). $\mathrm{ZF+sTD}$ implies that for any set of reals $A$ and any $ε>0$, (a) there is a closed set $F\subseteq A$ so that $\mathrm{Dim_H}(F)\geq \mathrm{Dim_H}(A)-ε$. (b) there is a closed set $F\subseteq A$ so that $\mathrm{Dim_P}(F)\geq \mathrm{Dim_P}(A)-ε$.

math.LO↗

arXiv2021

The TESS Objects of Interest Catalog from the TESS Prime Mission

We present 2,241 exoplanet candidates identified with data from the Transiting Exoplanet Survey Satellite (TESS) during its two-year prime mission. We list these candidates in the TESS Objects of Interest (TOI) Catalog, which includes both new planet candidates found by TESS and previously-known planets recovered by TESS observations. We describe the process used to identify TOIs and investigate the characteristics of the new planet candidates, and discuss some notable TESS planet discoveries. The TOI Catalog includes an unprecedented number of small planet candidates around nearby bright stars, which are well-suited for detailed follow-up observations. The TESS data products for the Prime Mission (Sectors 1-26), including the TOI Catalog, light curves, full-frame images, and target pixel files, are publicly available on the Mikulski Archive for Space Telescopes.

astro-ph.EP↗

arXiv2020

TD implies CCR

Assuming $\mathrm{ZF}$, we prove that Turing determinacy ($\mathrm{TD}$) implies countable choice axiom for sets of reals ($\mathrm{CCR}$).

math.LO↗