Open Access. Powered by Scholars. Published by Universities.®

Physical Sciences and Mathematics Commons™

Open Access. Powered by Scholars. Published by Universities.®

Artificial Intelligence and Robotics

Institution
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 1411 - 1440 of 8513

Full-Text Articles in Physical Sciences and Mathematics

Dqn-Based Joint Scheduling Method Of Heterogeneous Tt&C Resources, Naiyang Xue, Dan Ding, Yutong Jia, Zhiqiang Wang, Yuan Liu Feb 2023

Dqn-Based Joint Scheduling Method Of Heterogeneous Tt&C Resources, Naiyang Xue, Dan Ding, Yutong Jia, Zhiqiang Wang, Yuan Liu

Journal of System Simulation

Abstract: Joint scheduling of heterogeneous TT&C resources as research object, a deep Q network (DQN) algorithm based on reinforcement learning is proposed. The characteristics of the joint scheduling problem of heterogeneous TT&C resources being fully analyzied and mathematical language being used to describe the constraints affecting the solution, a resource joint scheduling model is established. From the perspective of applying reinforcement learning, two neural networks with the same structure and the action selection strategies based onεgreedy algorithm are respectively designed after Markov decision process description, and DQN solution framework is established. The simulation results show that DQN-based heterogeneous …

Go to article

Efficient Hmm Map Matching Method Using R-Tree And Trajectory Segmentation, Yanjiao Song, Jiayue Zhou, Longhao Wang, Jing Wu, Rui Li, Xiaoping Rui Feb 2023

Efficient Hmm Map Matching Method Using R-Tree And Trajectory Segmentation, Yanjiao Song, Jiayue Zhou, Longhao Wang, Jing Wu, Rui Li, Xiaoping Rui

Journal of System Simulation

Abstract: In view of the incapability of traditional methods to efficiently process massive trajectory data, an improved HMM (hidden-Markov model) map matching algorithm is proposed. Spatial index for road networks is established through R-tree spatial index. GPS trajectory data are segmented based on the position change rates of trajectory points. R-tree index is used to quickly determine the candidate road section that sub-trajectories belong to, and the key points of the sub-trajectories instead of the entire sub-trajectories are selected to judge which road the sub-trajectories should be matched with. The map matching of each sub-trajectory is carried out on …

Go to article

Research On Digital Twin Credibility Assessment Process And Index, Fan Yang, Ping Ma, Wei Li, Ming Yang Feb 2023

Research On Digital Twin Credibility Assessment Process And Index, Fan Yang, Ping Ma, Wei Li, Ming Yang

Journal of System Simulation

Abstract: With the application field expansion of digital twin technology, in order to meet the requirement of digital twin credibility and promote the digital twin credibility assessment, the credibility assessment process and indicators of digital twin are researched. The development process of digital twin is analyzed and the construction method of flexible multi-layer digital twin credibility assessment process model based on IDEF0 is proposed. Two index systems, process-stage-activity layers(P-S-A) and activity-element-feature layers(A-E-F),are proposed to solve the problems of defect backtracking and evaluation of complex objects. Several examples of index system are given.

Go to article

Online Classification Method For Motor Imagery Eeg With Spatial Information, Fengwei Yang, Peng Chen, Kai Xi, Hualin Pu, Xueyin Liu Feb 2023

Online Classification Method For Motor Imagery Eeg With Spatial Information, Fengwei Yang, Peng Chen, Kai Xi, Hualin Pu, Xueyin Liu

Journal of System Simulation

Abstract: EEG-based BCI system can help the daily life and rehabilitation training of limb movement disorders patients. Due to the low signal-to-noise ratio and large individual differences of EEG signals, the accuracy and efficiency of EEG feature extraction and classification are not high, which affects the wide application of online BCI system. A CNN with spatial information is proposed for the online classification of MI-EEG signals. The reordered MI-EEG is convolved horizontally and vertically respectively. With the contralateral effect of motor imagery ERD/ERS phenomenon, the spatial information in MI-EEG is fully utilized to achieve the real-time acquisition and classification of …

Go to article

Research On Image Super-Resolution Reconstruction Based On Loss Extraction Feedback Attention Network, Hong Sun, Yuxiang Zhang, Yuelan Ling Feb 2023

Research On Image Super-Resolution Reconstruction Based On Loss Extraction Feedback Attention Network, Hong Sun, Yuxiang Zhang, Yuelan Ling

Journal of System Simulation

Abstract: Since the first application of convolutional neural network to the field of super-resolution image reconstruction (super-resolution convolutional neural network, SRCNN), a large number of studies have proved that deep learning can improve the effect of image reconstruction. Aiming at the too many parameters in the image super-resolution network and the insufficient utilization of image features resulting in less available high-frequency information, a loss extraction feedback attention network (LEFAN) is proposed to reuse parameters in a circular way and increase the reuse of low-resolution image features to capture more high-frequency information. The loss caused in the reconstruction process is extracted …

Go to article

Hardware-In-The-Loop Simulation Platform Of Loop Control For Municipal Solid Waste Incineration Process, Tianzheng Wang, Jian Tang, Heng Xia, Junfei Qiao Feb 2023

Hardware-In-The-Loop Simulation Platform Of Loop Control For Municipal Solid Waste Incineration Process, Tianzheng Wang, Jian Tang, Heng Xia, Junfei Qiao

Journal of System Simulation

Abstract: To accurately simulate and realize the multiple input multiple output (MIMO) loop control of municipal solid waste incineration (MSWI) process, a distributed hardware-in-the-loop simulation platform consisting of a real device layer and a virtual object layer is developed based on the actual industrial process. The mechanism model is qualitatively described, and a data-driven virtual process object model in terms of loop control is established. The software subsystems of the platform and their cooperative operation mode are designed based on the control requirement. The hardware and software of the proposed platform are built and experimentally verified based on actual industrial …

Go to article

A Simulation Method Of Airborne Radar Real-Time Detection Based On Three-Dimensional Subdivision, Ying Xu, Shuai Zhang, Zhige Xie, Xinhai Xu, Manhui Sun, Ning Guo Feb 2023

A Simulation Method Of Airborne Radar Real-Time Detection Based On Three-Dimensional Subdivision, Ying Xu, Shuai Zhang, Zhige Xie, Xinhai Xu, Manhui Sun, Ning Guo

Journal of System Simulation

Abstract: The emergence and rapid development of UAVs make the target detection of UAV airborne radar in combat simulation great research valuable. In the existing combat simulation platforms at home and abroad, the detection relationship between radar and target is pairwise interactive, and the calculation overhead increases linearly or ultra-linear following the increase of entity numbers, which is difficult to carry out the large-scale real-time combat simulation. Based on the concept of three-dimensional meshing, an airborne radar target detection simulation method is proposed, which can quickly judge the success or failure of detection by making a detection template before simulation …

Go to article

Se(3)-Based Finite-Time Fault-Tolerant Control Of Spacecraft Integrated Attitude-Orbit, Yafei Mei, Ying Liao, Kejie Gong, Xingyu Zheng Feb 2023

Se(3)-Based Finite-Time Fault-Tolerant Control Of Spacecraft Integrated Attitude-Orbit, Yafei Mei, Ying Liao, Kejie Gong, Xingyu Zheng

Journal of System Simulation

Abstract: For relative motion spacecraft, when the actuator fails and the external disturbance and system uncertainty occur simultaneously, a finite-time fault-tolerant control method is proposed on the basis of the robustness of sliding mode control. A single-rigid spacecraft integrated attitude-orbit model is established based on Lie Group SE(3), and the relative motion spacecraft error dynamic equation is derived in exponential coordinates. A class of non-singular fast terminal sliding surface is designed, and the equivalent adaptive method is adopted to design the controller to estimate and compensate the total disturbance. The proposed fault-tolerant control algorithm can be independent from fault diagnosis …

Go to article

Simulation Of Occluded Pedestrian Detection Based On Improved Yolo, Nan Xiang, Lu Wang, Chongliu Jia, Yuemou Jian, Xiaoxia Ma Feb 2023

Simulation Of Occluded Pedestrian Detection Based On Improved Yolo, Nan Xiang, Lu Wang, Chongliu Jia, Yuemou Jian, Xiaoxia Ma

Journal of System Simulation

Abstract: Aiming at the high missed detection rates and low accuracy of existing YOLO for occlusion and multi-scale pedestrian targets, an improved pedestrian detection algorithm is proposed. YOLO backbone is modified to enhance the capabilities of cross-scale feature extraction. To increase thepedestrian feature fusion capabilities of different scales, a spatial pyramid pooling module and two attention mechanisms are introduced at different positions in front of YOLO layers. Aiming at the detection performance degradation due to the extreme complexity of network module and to improve the model training efficiency, the network structure is pruned according to the actual …

Go to article

Evacuation Dynamics Research Based On Evolutionary Game Theory, Qiaoru Li, Jinxiu Yan, Xiaoyong Tian, Kun Li, Xia Li Feb 2023

Evacuation Dynamics Research Based On Evolutionary Game Theory, Qiaoru Li, Jinxiu Yan, Xiaoyong Tian, Kun Li, Xia Li

Journal of System Simulation

Abstract: In order to study the impact of individual cooperative behavior on the overall pedestrian evacuation efficiency, combined with cellular automata and social force model, the co-evolution of evacuation system dynamics and "cooperative behavior" is studied, and an evacuation dynamics research method based on evolutionary game theory is proposed. The cellular automata model is used as the basic simulation framework, and the social force model is used to represent the psychological repulsion among the practitioners. The evacuation individual strategy is updated through evolutionary game. The simulation results show that when the game gain coefficient exceeds a certain threshold, the …

Go to article

Design And Implementation Of Industrial Robot Remote Monitoring System In Cloud Manufacturing, Yongkui Liu, Lin Zhang, Yingfu Liu, Jianyong Feng, Bo Yu, Wenbo Niu Feb 2023

Design And Implementation Of Industrial Robot Remote Monitoring System In Cloud Manufacturing, Yongkui Liu, Lin Zhang, Yingfu Liu, Jianyong Feng, Bo Yu, Wenbo Niu

Journal of System Simulation

Abstract: Considering the lack of the existing research on cloud manufacturing monitoring system and the lack of scalability and flexibility of existing remote monitoring system, and taking deep reinforcement learning-based industrial robot intelligent grasping as an application scenario, a micro-service architecture-based remote monitoring system for cloud manufacturing is developed, to carry out the requirement analysis and design of the monitoring system, and the remote monitoring of industrial robot intelligent grasping processes is realized. Test results show that the system can meet the monitoring requirements of resource providers, platform operator(s) and service consumers.

Go to article

Design Of System Combat Simulation Platform For Complex Electromagnetic Environment, Baiyuan Ding, Fuling Mu, Yunpeng Li, Zhongkuan Chen, Chengyu Liu Feb 2023

Design Of System Combat Simulation Platform For Complex Electromagnetic Environment, Baiyuan Ding, Fuling Mu, Yunpeng Li, Zhongkuan Chen, Chengyu Liu

Journal of System Simulation

Abstract: War gaming and simulation can be divided into four levels of strategy、campaign、tactics and technique. Existing system combat simulation platforms of campaign and tactics at home and abroad cannot drive the special model of electromagnetic equipment in technique level, resulting in the low fidelity of battlefield complex electromagnetic environment simulation. To solve the problem, flexible analysis modeling and exercise system(FLAMES) model architecture as the reference, based on the simulation engine library of a discrete event system simulator(ADEVS), a flexible operation simulation platform(FOSim) is designed and developed, which integrates three levels of campaign, tactics and technology, supports three levels independence simulation …

Go to article

Machine Learning-Based Simulation Research Of On-Line Subway Pedestrian Flow Control, Jiajie Shi, Peng Yang, Yannan Pi Feb 2023

Machine Learning-Based Simulation Research Of On-Line Subway Pedestrian Flow Control, Jiajie Shi, Peng Yang, Yannan Pi

Journal of System Simulation

Abstract: In recent years, a large number of digital experiments have been carried out in the field of space launch, such as digital design verification, digital joint training and simulation training, and rocket-ground joint simulation evaluation, all of which involve the space launch information visualization. Through virtual reality technology, system simulation technology, data visualization technology, etc., on the basis of multi-thread, multi-module architecture design idea, and message queue system interaction mode, the space launch visual simulation analysis technology platform with functions of data management, scenario management, calculation management, and script management is constructed. The application cases of visual simulation analysis …

Go to article

Takeout Distribution Routes Optimization Considering Order Clustering Under Dynamic Demand, Houming Fan, Fushan Xian, Huaiqi Wang Feb 2023

Takeout Distribution Routes Optimization Considering Order Clustering Under Dynamic Demand, Houming Fan, Fushan Xian, Huaiqi Wang

Journal of System Simulation

Abstract: Takeout distribution optimization includes order allocation and route planning. Aiming at dynamic order and rider position change, with the goal of minimizing the overtime order proportion, average delivery time and average travel distance,a two-stage mathematical model is established based on the idea of pre-optimization and dynamic adjustment. In the pre-optimization stage, an improved variable neighborhood search algorithm is designed to obtain the initial distribution scheme. In the dynamic adjustment stage, a periodic optimization strategy is adopted to transform the problem into a virtual distribution center vehicle problem for solution. In each stage,different clustering methods are used to optimize the …

Go to article

Research On Cooperative Path Planning Model Of Multiple Unmanned Vehicles In Real Environment, Guohui Zhang, Xuan Wang, Yanan Zhang, Ang Gao Feb 2023

Research On Cooperative Path Planning Model Of Multiple Unmanned Vehicles In Real Environment, Guohui Zhang, Xuan Wang, Yanan Zhang, Ang Gao

Journal of System Simulation

Abstract: The cluster combat application of unmanned ground vehicles(UVS) is a hot research issue of the intersection of artificial intelligence and battle command. Aiming at the cooperative path planning multiple unmanned vehicles not meeting the dynamic threat condition requirement, by combining the global path planning algorithm A-STAR with the local path planning algorithm RL, from the perspective of perception to behavioral decision making, the cooperative path planning model of multiple unmanned vehicles is studied. The cooperative combat situation threat algorithm, state and action space, reward function and sphere of influence function are designed, the sub-models of formation configuration strategy generation …

Go to article

Machine Learning Methods For Computational Phenotyping Using Patient Healthcare Data With Noisy Labels, Praveen Kumar Feb 2023

Machine Learning Methods For Computational Phenotyping Using Patient Healthcare Data With Noisy Labels, Praveen Kumar

Computer Science ETDs

Positive and Unlabeled (PU) learning problems abound in many real-world applications. In healthcare informatics, diagnosed patients are considered labeled positive for a specific disease, but being undiagnosed does not mean they can be labeled negative. PU learning can improve classification performance, and estimate the positive fraction, α, among unlabeled samples. However, algorithms based on the Selected Completely At Random (SCAR) assumption are inadequate when the SCAR assumption fails (e.g., severe cases overrepresented), and when class imbalance is substantial. This dissertation presents and evaluates new algorithms to overcome these limitations. The proposed methods outperform the state-of-art for α-estimation, enhance classification performance, …

Go to article

Session11: Skip-Gcn : A Framework For Hierarchical Graph Representation Learning, Jackson Cates, Justin Lewis, Randy Hoover, Kyle Caudle Feb 2023

Session11: Skip-Gcn : A Framework For Hierarchical Graph Representation Learning, Jackson Cates, Justin Lewis, Randy Hoover, Kyle Caudle

SDSU Data Science Symposium

Recently there has been high demand for the representation learning of graphs. Graphs are a complex data structure that contains both topology and features. There are first several domains for graphs, such as infectious disease contact tracing and social media network communications interactions. The literature describes several methods developed that work to represent nodes in an embedding space, allowing for classical techniques to perform node classification and prediction. One such method is the graph convolutional neural network that aggregates the node neighbor’s features to create the embedding. Another method, Walklets, takes advantage of the topological information stored in a graph …

Go to article

Temporal Tensor Factorization For Multidimensional Forecasting, Jackson Cates, Karissa Scipke, Randy Hoover, Kyle Caudle Feb 2023

Temporal Tensor Factorization For Multidimensional Forecasting, Jackson Cates, Karissa Scipke, Randy Hoover, Kyle Caudle

SDSU Data Science Symposium

In the era of big data, there is a need for forecasting high-dimensional time series that might be incomplete, sparse, and/or nonstationary. The current research aims to solve this problem for two-dimensional data through a combination of temporal matrix factorization (TMF) and low-rank tensor factorization. From this method, we propose an expansion of TMF to two-dimensional data: temporal tensor factorization (TTF). The current research aims to interpolate missing values via low-rank tensor factorization, which produces a latent space of the original multilinear time series. We then can perform forecasting in the latent space. We present experimental results of the proposed …

Go to article

A Proposed Meta-Reality Immersive Development Pipeline: Generative Ai Models And Extended Reality (Xr) Content For The Metaverse, Jay Ratican, James Hutson, Andrew Wright Feb 2023

A Proposed Meta-Reality Immersive Development Pipeline: Generative Ai Models And Extended Reality (Xr) Content For The Metaverse, Jay Ratican, James Hutson, Andrew Wright

Faculty Scholarship

The realization of an interoperable and scalable virtual platform, currently known as the “metaverse,” is inevitable, but many technological challenges need to be overcome first. With the metaverse still in a nascent phase, research currently indicates that building a new 3D social environment capable of interoperable avatars and digital transactions will represent most of the initial investment in time and capital. The return on investment, however, is worth the financial risk for firms like Meta, Google, and Apple. While the current virtual space of the metaverse is worth $6.30 billion, that is expected to grow to $84.09 billion by the …

Go to article

Emotion Classification Of Indonesian Tweets Using Bidirectional Lstm, Aaron K. Glenn, Phillip M. Lacasse, Bruce A. Cox Feb 2023

Emotion Classification Of Indonesian Tweets Using Bidirectional Lstm, Aaron K. Glenn, Phillip M. Lacasse, Bruce A. Cox

Faculty Publications

Emotion classification can be a powerful tool to derive narratives from social media data. Traditional machine learning models that perform emotion classification on Indonesian Twitter data exist but rely on closed-source features. Recurrent neural networks can meet or exceed the performance of state-of-the-art traditional machine learning techniques using exclusively open-source data and models. Specifically, these results show that recurrent neural network variants can produce more than an 8% gain in accuracy in comparison with logistic regression and SVM techniques and a 15% gain over random forest when using FastText embeddings. This research found a statistical significance in the performance of …

Go to article

Towards Carbon Neutrality: Prediction Of Wave Energy Based On Improved Gru In Maritime Transportation, Zhihan Lv, Nana Wang, Ranran Lou, Yajun Tian, Mohsen Guizani Feb 2023

Towards Carbon Neutrality: Prediction Of Wave Energy Based On Improved Gru In Maritime Transportation, Zhihan Lv, Nana Wang, Ranran Lou, Yajun Tian, Mohsen Guizani

Machine Learning Faculty Publications

Efficient use of renewable energy is one of the critical measures to achieve carbon neutrality. Countries have introduced policies to put carbon neutrality on the agenda to achieve relatively zero emissions of greenhouse gases and to cope with the crisis brought about by global warming. This work analyzes the wave energy with high energy density and wide distribution based on understanding of various renewable energy sources. This study provides a wave energy prediction model for energy harvesting. At the same time, the Gated Recurrent Unit network (GRU), Bayesian optimization algorithm, and attention mechanism are introduced to improve the model's performance. …

Go to article

Drone Detection Using Yolov5, Burchan Aydin, Subroto Singha Feb 2023

Drone Detection Using Yolov5, Burchan Aydin, Subroto Singha

Faculty Publications

The rapidly increasing number of drones in the national airspace, including those for recreational and commercial applications, has raised concerns regarding misuse. Autonomous drone detection systems offer a probable solution to overcoming the issue of potential drone misuse, such as drug smuggling, violating people’s privacy, etc. Detecting drones can be difficult, due to similar objects in the sky, such as airplanes and birds. In addition, automated drone detection systems need to be trained with ample amounts of data to provide high accuracy. Real-time detection is also necessary, but this requires highly configured devices such as a graphical processing unit (GPU). …

Go to article

Real-Time Hierarchical Map Segmentation For Coordinating Multi-Robot Exploration, Tianze Luo, Zichen Chen, Budhitama Subagdja, Ah-Hwee Tan Feb 2023

Real-Time Hierarchical Map Segmentation For Coordinating Multi-Robot Exploration, Tianze Luo, Zichen Chen, Budhitama Subagdja, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Coordinating a team of autonomous agents to explore an environment can be done by partitioning the map of the environment into segments and allocating the segments as targets for the individual agents to visit. However, given an unknown environment, map segmentation must be conducted in a continuous and incremental manner. In this paper, we propose a novel real-time hierarchical map segmentation method for supporting multi-agent exploration of indoor environments, wherein clusters of regions of segments are formed hierarchically from randomly sampled points in the environment. Each cluster is then assigned with a cost-utility value based on the minimum cost possible …

Go to article

Uncertaintyfusenet: Robust Uncertainty-Aware Hierarchical Feature Fusion Model With Ensemble Monte Carlo Dropout For Covid-19 Detection, Moloud Abdar, Soorena Salari, Sina Qahremani, Hak-Keung Lam, Fakhreddine (Fakhri) Karray, Sadiq Hussain, Abbas Khosravi, U. Rajendra Acharya, Vladimir Makarenkov, Saeid Nahavandi Feb 2023

Uncertaintyfusenet: Robust Uncertainty-Aware Hierarchical Feature Fusion Model With Ensemble Monte Carlo Dropout For Covid-19 Detection, Moloud Abdar, Soorena Salari, Sina Qahremani, Hak-Keung Lam, Fakhreddine (Fakhri) Karray, Sadiq Hussain, Abbas Khosravi, U. Rajendra Acharya, Vladimir Makarenkov, Saeid Nahavandi

Machine Learning Faculty Publications

The COVID-19 (Coronavirus disease 2019) pandemic has become a major global threat to human health and well-being Thus, the development of computer-aided detection (CAD) systems that are capable to accurately distinguish COVID-19 from other diseases using chest computed tomography (CT) and X-ray data is of immediate priority Such automatic systems are usually based on traditional machine learning or deep learning methods Differently from most of existing studies, which used either CT scan or X-ray images in COVID-19-case classification, we present a simple but efficient deep learning feature fusion model, called UncertaintyFuseNet, which is able to classify accurately large datasets of …

Go to article

Scalable And Globally Optimal Generalized L1 K-Center Clustering Via Constraint Generation In Mixed Integer Linear Programming, Aravinth Chembu, Scott Sanner, Hassan Khurran, Akshat Kumar Feb 2023

Scalable And Globally Optimal Generalized L1 K-Center Clustering Via Constraint Generation In Mixed Integer Linear Programming, Aravinth Chembu, Scott Sanner, Hassan Khurran, Akshat Kumar

Research Collection School Of Computing and Information Systems

The k-center clustering algorithm, introduced over 35 years ago, is known to be robust to class imbalance prevalent in many clustering problems and has various applications such as data summarization, document clustering, and facility location determination. Unfortunately, existing k-center algorithms provide highly suboptimal solutions that can limit their practical application, reproducibility, and clustering quality. In this paper, we provide a novel scalable and globally optimal solution to a popular variant of the k-center problem known as generalized L_1 k-center clustering that uses L_1 distance and allows the selection of arbitrary vectors as cluster centers. We show that this clustering objective …

Go to article

Legal Dispositionism And Artificially-Intelligent Attributions, Jerrold Soh Feb 2023

Legal Dispositionism And Artificially-Intelligent Attributions, Jerrold Soh

Research Collection Yong Pung How School Of Law

It is conventionally argued that because an artificially-intelligent (AI) system acts autonomously, its makers cannot easily be held liable should the system's actions harm. Since the system cannot be liable on its own account either, existing laws expose victims to accountability gaps and need to be reformed. Recent legal instruments have nonetheless established obligations against AI developers and providers. Drawing on attribution theory, this paper examines how these seemingly opposing positions are shaped by the ways in which AI systems are conceptualised. Specifically, folk dispositionism underpins conventional legal discourse on AI liability, personality, publications, and inventions and leads us towards …

Go to article

Online Hyperparameter Optimization For Class-Incremental Learning, Yaoyao Liu, Yingying Li, Bernt Schiele, Qianru Sun Feb 2023

Online Hyperparameter Optimization For Class-Incremental Learning, Yaoyao Liu, Yingying Li, Bernt Schiele, Qianru Sun

Research Collection School Of Computing and Information Systems

Class-incremental learning (CIL) aims to train a classification model while the number of classes increases phase-by-phase. An inherent challenge of CIL is the stability-plasticity tradeoff, i.e., CIL models should keep stable to retain old knowledge and keep plastic to absorb new knowledge. However, none of the existing CIL models can achieve the optimal tradeoff in different data-receiving settings—where typically the training-from-half (TFH) setting needs more stability, but the training-from-scratch (TFS) needs more plasticity. To this end, we design an online learning method that can adaptively optimize the tradeoff without knowing the setting as a priori. Specifically, we first introduce the …

Go to article

Generalization Bounds For Inductive Matrix Completion In Low-Noise Settings, Antoine Ledent, Rodrigo Alves, Yunwen Lei, Yann Guermeur, Marius Kloft Feb 2023

Generalization Bounds For Inductive Matrix Completion In Low-Noise Settings, Antoine Ledent, Rodrigo Alves, Yunwen Lei, Yann Guermeur, Marius Kloft

Research Collection School Of Computing and Information Systems

We study inductive matrix completion (matrix completion with side information) under an i.i.d. subgaussian noise assumption at a low noise regime, with uniform sampling of the entries. We obtain for the first time generalization bounds with the following three properties: (1) they scale like the standard deviation of the noise and in particular approach zero in the exact recovery case; (2) even in the presence of noise, they converge to zero when the sample size approaches infinity; and (3) for a fixed dimension of the side information, they only have a logarithmic dependence on the size of the matrix. Differently …

Go to article

Layout Generation As Intermediate Action Sequence Prediction, Huiting Yang, Danqing Huang, Chin-Yew Lin, Shengfeng He Feb 2023

Layout Generation As Intermediate Action Sequence Prediction, Huiting Yang, Danqing Huang, Chin-Yew Lin, Shengfeng He

Research Collection School Of Computing and Information Systems

Layout generation plays a crucial role in graphic design intelligence. One important characteristic of the graphic layouts is that they usually follow certain design principles. For example, the principle of repetition emphasizes the reuse of similar visual elements throughout the design. To generate a layout, previous works mainly attempt at predicting the absolute value of bounding box for each element, where such target representation has hidden the information of higher-order design operations like repetition (e.g. copy the size of the previously generated element). In this paper, we introduce a novel action schema to encode these operations for better modeling the …

Go to article

Generalizing Math Word Problem Solvers Via Solution Diversification, Zhenwen Liang, Jipeng Zhang, Lei Wang, Yan Wang, Jie Shao, Xiangliang Zhang Feb 2023

Generalizing Math Word Problem Solvers Via Solution Diversification, Zhenwen Liang, Jipeng Zhang, Lei Wang, Yan Wang, Jie Shao, Xiangliang Zhang

Research Collection School Of Computing and Information Systems

Current math word problem (MWP) solvers are usually Seq2Seq models trained by the (one-problem; one-solution) pairs, each of which is made of a problem description and a solution showing reasoning flow to get the correct answer. However, one MWP problem naturally has multiple solution equations. The training of an MWP solver with (one-problem; one-solution) pairs excludes other correct solutions, and thus limits the generalizability of the MWP solver. One feasible solution to this limitation is to augment multiple solutions to a given problem. However, it is difficult to collect diverse and accurate augment solutions through human efforts. In this paper, …

Go to article

First Prev Next Last