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

Computer Sciences Commons™

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

Artificial Intelligence and Robotics

Institution
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 5101 - 5130 of 11355

Full-Text Articles in Computer Sciences

Reinforcement-Learning-Based Adaptive Tracking Control For A Space Continuum Robot Based On Reinforcement Learning, Da Jiang, Zhiqin Cai, Zhongzhen Liu, Haijun Peng, Zhigang Wu Oct 2022

Reinforcement-Learning-Based Adaptive Tracking Control For A Space Continuum Robot Based On Reinforcement Learning, Da Jiang, Zhiqin Cai, Zhongzhen Liu, Haijun Peng, Zhigang Wu

Journal of System Simulation

Abstract: Aiming at the tracking control for three-arm space continuum robot in space active debris removal manipulation, an adaptive sliding mode control algorithm based on deep reinforcement learning is proposed. Through BP network, a data-driven dynamic model is developed as the predictive model to guide the reinforcement learning to adjust the sliding mode controller's parameters online, and finally realize a real-time tracking control. Simulation results show that the proposed data-driven predictive model can accurately predict the robot's dynamic characteristics with the relative error within ±1% to random trajectories. Compared with the fixed-parameter sliding mode controller, the proposed adaptive controller …

Go to article

Simulation Research On Aerodynamic Characteristics Of A Miniature Munition, Haisen Wang, Junfang Fan Oct 2022

Simulation Research On Aerodynamic Characteristics Of A Miniature Munition, Haisen Wang, Junfang Fan

Journal of System Simulation

Abstract: Aiming at the existing constraints of layout design of miniature munition, a new layout optimization scheme is proposed. Based on the normal layout structure, a three-dimensional aerodynamic model of munition is established, and the aerodynamic hydrodynamics simulation of miniature munition is carried out, and the influence of wing size on static stability of munition is studied. To verify the feasibility of the design scheme, the aerodynamic simulation calculation is carried out, and the lift, drag and pitching moment parameters of miniature munition are obtained through simulated aerodynamic calculation under different flight conditions, and the lift-drag ratio …

Go to article

Fatigue Detection Method Based On Facial Features And Head Posture, Rongxiu Lu, Bihao Zhang, Zhenlong Mo Oct 2022

Fatigue Detection Method Based On Facial Features And Head Posture, Rongxiu Lu, Bihao Zhang, Zhenlong Mo

Journal of System Simulation

Abstract: Aiming at the of the single fatigue characteristics, low robustness and inability to customize fatigue thresholds for different drivers of fatigue detection methods, a method based on facial features and head posture is proposed. In face detection and face key point positioning HOG feature operator and regression tree algorithm are used. In head posture estimation, head posture Euler angle is estimated by combining the face key points with the coordinate system transformation. In fatigue feature extraction, a deep residual neural network model is established to extract the eye fatigue features, which the eye, mouth aspect ratio and head posture …

Go to article

Task Allocation Method For Multi-Uav Cooperative Reconnaissance In Complex Environment, Fuzhen Zhang, Yaoqin Zhu Oct 2022

Task Allocation Method For Multi-Uav Cooperative Reconnaissance In Complex Environment, Fuzhen Zhang, Yaoqin Zhu

Journal of System Simulation

Abstract: The existing cooperative planning methods of multiple UAVs often carry out path planning and task allocation separatly, which causes the cooperative scheme not being the best in a complex environment. The cost matrix of multi-UAV cooperative reconnaissance on heterogeneous targets is established. Aiming at the various obstacle constraints and the characteristics of UAV motion and track in complex environment, an improved PSO-AFSA is used to solve the single UAV track planning model. Hungarian algorithm is used to complete the cooperative allocation of reconnaissance tasks of UAVs. The simulation results show that the algorithm can make the flying range of …

Go to article

Simulation Of Emergency Medical Materials Collaborative Distribution Considering Supplier Clustering, Zhe Wang, Hongyuan Shao, Zihao Cong, Wenwen Ma Oct 2022

Simulation Of Emergency Medical Materials Collaborative Distribution Considering Supplier Clustering, Zhe Wang, Hongyuan Shao, Zihao Cong, Wenwen Ma

Journal of System Simulation

Abstract: Aiming at the insufficient local government medical material reserves and low distribution efficiency to public health emergencies, consider supplier clustering a two-stage emergency medical supplies public-private collaborative location and alloation model is designed. In for disaster preparedness, fuzzy clustering algorithm to realize supplier clustering, and a cooperation mechanism with the government is establish realize the joint storage of emergency medical supplies; In the early stage of rescue, relying on the government and various suppliers' material storage and transportation capabilities, a location allocation simulation model to minimize the weighted sum of total logistics cost and demand unsatisfied rate is …

Go to article

Research On High-Performance Emulation Technology Of Starlink Constellation Based On Cloud Platform, Yuan Liu, Xinyi Xue, Xiaofeng Wang Oct 2022

Research On High-Performance Emulation Technology Of Starlink Constellation Based On Cloud Platform, Yuan Liu, Xinyi Xue, Xiaofeng Wang

Journal of System Simulation

Abstract: Network emulation on Starlink constellation is an important verification and evaluation tool for the design and construction of low earth orbit constellations in the future. Aiming at the characteristics of large-scale and complex structure of the Starlink network, a high-performance satellite network emulation system is designed. Based on the distributed network emulation architecture of cloud platform, through the development of STK Engine underlying interface, satellite model library storage optimization and asynchronous message transmission technology, the rapid deployment of Starlink constellation is carried out, and has good scalability. The experimental results show that the proposed method can carry out …

Go to article

Analysis And Research On End-To-End Optical Image Quality Of Large Aperture Off-Axis Space Telescope, Zhang Ban, Xiaobo Li, Xun Yang, Yuxi Jiang Oct 2022

Analysis And Research On End-To-End Optical Image Quality Of Large Aperture Off-Axis Space Telescope, Zhang Ban, Xiaobo Li, Xun Yang, Yuxi Jiang

Journal of System Simulation

Abstract: In order to evaluate the image quality of space on orbit telescope under the comprehensive constraints, the chain calculation analysis method is adopted. The main error factors of the optical image quality degradation are divided into six static errors and two dynamic errors. After calculation, the optical system wavefront aberration under the static loading error is reduced to 0.057λ on average. Under the influence of the static and dynamic errors, the 80% energy concentration of point spread function increases gradually. The method can be used to analyze and evaluate the influence of image stabilization control and precise temperature …

Go to article

Agglomerative Hierarchical Clustering With Dynamic Time Warping For Household Load Curve Clustering, Fadi Almahamid, Katarina Grolinger Oct 2022

Agglomerative Hierarchical Clustering With Dynamic Time Warping For Household Load Curve Clustering, Fadi Almahamid, Katarina Grolinger

Electrical and Computer Engineering Publications

Energy companies often implement various demand response (DR) programs to better match electricity demand and supply by offering the consumers incentives to reduce their demand during critical periods. Classifying clients according to their consumption patterns enables targeting specific groups of consumers for DR. Traditional clustering algorithms use standard distance measurement to find the distance between two points. The results produced by clustering algorithms such as K-means, K-medoids, and Gaussian Mixture Models depend on the clustering parameters or initial clusters. In contrast, our methodology uses a shape-based approach that combines Agglomerative Hierarchical Clustering (AHC) with Dynamic Time Warping (DTW) to classify …

Go to article

Tutorial: Neuro-Symbolic Ai For Mental Healthcare, Kaushik Roy, Usha Lokala, Manas Gaur, Amit Sheth Oct 2022

Tutorial: Neuro-Symbolic Ai For Mental Healthcare, Kaushik Roy, Usha Lokala, Manas Gaur, Amit Sheth

Publications

Artificial Intelligence (AI) systems for mental healthcare (MHCare) have been ever-growing after realizing the importance of early interventions for patients with chronic mental health (MH) conditions. Social media (SocMedia) emerged as the go-to platform for supporting patients seeking MHCare. The creation of peer-support groups without social stigma has resulted in patients transitioning from clinical settings to SocMedia supported interactions for quick help. Researchers started exploring SocMedia content in search of cues that showcase correlation or causation between different MH conditions to design better interventional strategies. User-level Classification-based AI systems were designed to leverage diverse SocMedia data from various MH conditions, …

Go to article

An Investigation Of The Reconstruction Capacity Of Stacked Convolutional Autoencoders For Log-Mel-Spectrograms, Anastasia Natsiou, Luca Longo, Seán O'Leary Oct 2022

An Investigation Of The Reconstruction Capacity Of Stacked Convolutional Autoencoders For Log-Mel-Spectrograms, Anastasia Natsiou, Luca Longo, Seán O'Leary

Conference Papers

In audio processing applications, the generation of expressive sounds based on high-level representations demonstrates a high demand. These representations can be used to manipulate the timbre and influence the synthesis of creative instrumental notes. Modern algorithms, such as neural networks, have inspired the development of expressive synthesizers based on musical instrument timbre compression. Unsupervised deep learning methods can achieve audio compression by training the network to learn a mapping from waveforms or spectrograms to low-dimensional representations. This study investigates the use of stacked convolutional autoencoders for the compression of time-frequency audio representations for a variety of instruments for a single …

Go to article

Visual Object Tracking With Discriminative Filters And Siamese Networks: A Survey And Outlook, Sajid Javed, Martin Danelljan, Fahad Shahbaz Khan, Muhammad Haris Khan, Michael Felsberg, Jiri Matas Oct 2022

Visual Object Tracking With Discriminative Filters And Siamese Networks: A Survey And Outlook, Sajid Javed, Martin Danelljan, Fahad Shahbaz Khan, Muhammad Haris Khan, Michael Felsberg, Jiri Matas

Computer Vision Faculty Publications

Accurate and robust visual object tracking is one of the most challenging and fundamental computer vision problems. It entails estimating the trajectory of the target in an image sequence, given only its initial location, and segmentation, or its rough approximation in the form of a bounding box. Discriminative Correlation Filters (DCFs) and deep Siamese Networks (SNs) have emerged as dominating tracking paradigms, which have led to significant progress. Following the rapid evolution of visual object tracking in the last decade, this survey presents a systematic and thorough review of more than 90 DCFs and Siamese trackers, based on results in …

Go to article

Ps-Arm: An End-To-End Attention-Aware Relation Mixer Network For Person Search, Mustansar Fiaz, Hisham Cholakkal, Sanath Narayan, Rao Anwer, Fahad Shahbaz Khan Oct 2022

Ps-Arm: An End-To-End Attention-Aware Relation Mixer Network For Person Search, Mustansar Fiaz, Hisham Cholakkal, Sanath Narayan, Rao Anwer, Fahad Shahbaz Khan

Computer Vision Faculty Publications

Person search is a challenging problem with various real-world applications, that aims at joint person detection and re-identification of a query person from uncropped gallery images. Although, previous study focuses on rich feature information learning, it’s still hard to retrieve the query person due to the occurrence of appearance deformations and background distractors. In this paper, we propose a novel attention-aware relation mixer (ARM) module for person search, which exploits the global relation between different local regions within RoI of a person and make it robust against various appearance deformations and occlusion. The proposed ARM is composed of a relation …

Go to article

Artificial Intelligence And The Situational Rationality Of Diagnosis: Human Problem-Solving And The Artifacts Of Health And Medicine, Michael W. Raphael Oct 2022

Artificial Intelligence And The Situational Rationality Of Diagnosis: Human Problem-Solving And The Artifacts Of Health And Medicine, Michael W. Raphael

Publications and Research

What is the problem-solving capacity of artificial intelligence (AI) for health and medicine? This paper draws out the cognitive sociological context of diagnostic problem-solving for medical sociology regarding the limits of automation for decision-based medical tasks. Specifically, it presents a practical way of evaluating the artificiality of symptoms and signs in medical encounters, with an emphasis on the visualization of the problem-solving process in doctor-patient relationships. In doing so, the paper details the logical differences underlying diagnostic task performance between man and machine problem-solving: its principle of rationality, the priorities of its means of adaptation to abstraction, and the effects …

Go to article

Hyperfast Second-Order Local Solvers For Efficient Statistically Preconditioned Distributed Optimization, Pavel Dvurechensky, Dmitry Kamzolov, Aleksandr Lukashevich, Soomin Lee, Erik Ordentlich, César A. Uribe, Alexander Gasnikov Oct 2022

Hyperfast Second-Order Local Solvers For Efficient Statistically Preconditioned Distributed Optimization, Pavel Dvurechensky, Dmitry Kamzolov, Aleksandr Lukashevich, Soomin Lee, Erik Ordentlich, César A. Uribe, Alexander Gasnikov

Machine Learning Faculty Publications

Statistical preconditioning enables fast methods for distributed large-scale empirical risk minimization problems. In this approach, multiple worker nodes compute gradients in parallel, which are then used by the central node to update the parameter by solving an auxiliary (preconditioned) smaller-scale optimization problem. The recently proposed Statistically Preconditioned Accelerated Gradient (SPAG) method [1] has complexity bounds superior to other such algorithms but requires an exact solution for computationally intensive auxiliary optimization problems at every iteration. In this paper, we propose an Inexact SPAG (InSPAG) and explicitly characterize the accuracy by which the corresponding auxiliary subproblem needs to be solved to guarantee …

Go to article

A Survey On Multimodal Disinformation Detection, Firoj Alam, Stefano Cresci, Tanmoy Chakraborty, Fabrizio Silvestri, Dimitar Dimitrov, Giovanni Da San Martino, Shaden Shaar, Hamed Firooz, Preslav Nakov Oct 2022

A Survey On Multimodal Disinformation Detection, Firoj Alam, Stefano Cresci, Tanmoy Chakraborty, Fabrizio Silvestri, Dimitar Dimitrov, Giovanni Da San Martino, Shaden Shaar, Hamed Firooz, Preslav Nakov

Natural Language Processing Faculty Publications

Recent years have witnessed the proliferation of offensive content online such as fake news, propaganda, misinformation, and disinformation. While initially this was mostly about textual content, over time images and videos gained popularity, as they are much easier to consume, attract more attention, and spread further than text. As a result, researchers started leveraging different modalities and combinations thereof to tackle online multimodal offensive content. In this study, we offer a survey on the state-of-the-art on multimodal disinformation detection covering various combinations of modalities: text, images, speech, video, social media network structure, and temporal information. Moreover, while some studies focused …

Go to article

Noisy Label Regularisation For Textual Regression, Yuxia Wang, Timothy Baldwin, Karin Verspoor Oct 2022

Noisy Label Regularisation For Textual Regression, Yuxia Wang, Timothy Baldwin, Karin Verspoor

Natural Language Processing Faculty Publications

Training with noisy labelled data is known to be detrimental to model performance, especially for high-capacity neural network models in low-resource domains. Our experiments suggest that standard regularisation strategies, such as weight decay and dropout, are ineffective in the face of noisy labels. We propose a simple noisy label detection method that prevents error propagation from the input layer. The approach is based on the observation that the projection of noisy labels is learned through memorisation at advanced stages of learning, and that the Pearson correlation is sensitive to outliers. Extensive experiments over real-world human-disagreement annotations as well as randomly-corrupted …

Go to article

Lemurs Optimizer: A New Metaheuristic Algorithm For Global Optimization, Ammar Kamal Abasi, Sharif Naser Makhadmeh, Mohammed Azmi Al-Betar, Osama Ahmad Alomari, Mohammed A. Awadallah, Zaid Abdi Alkareem Alyasseri, Iyad Abu Doush, Ashraf Elnagar, Eman H. Alkhammash, Myriam Hadjouni Oct 2022

Lemurs Optimizer: A New Metaheuristic Algorithm For Global Optimization, Ammar Kamal Abasi, Sharif Naser Makhadmeh, Mohammed Azmi Al-Betar, Osama Ahmad Alomari, Mohammed A. Awadallah, Zaid Abdi Alkareem Alyasseri, Iyad Abu Doush, Ashraf Elnagar, Eman H. Alkhammash, Myriam Hadjouni

Machine Learning Faculty Publications

The Lemur Optimizer (LO) is a novel nature-inspired algorithm we propose in this paper. This algorithm’s primary inspirations are based on two pillars of lemur behavior: leap up and dance hub. These two principles are mathematically modeled in the optimization context to handle local search, exploitation, and exploration search concepts. The LO is first benchmarked on twenty-three standard optimization functions. Additionally, the LO is used to solve three real-world problems to evaluate its performance and effectiveness. In this direction, LO is compared to six well-known algorithms: Salp Swarm Algorithm (SSA), Artificial Bee Colony (ABC), Sine Cosine Algorithm (SCA), Bat Algorithm …

Go to article

Transrepair: Context-Aware Program Repair For Compilation Errors, Xueyang Li, Shangqing Liu, Ruitao Feng, Guozhu Meng, Xiaofei Xie, Kai Chen, Yang Liu Oct 2022

Transrepair: Context-Aware Program Repair For Compilation Errors, Xueyang Li, Shangqing Liu, Ruitao Feng, Guozhu Meng, Xiaofei Xie, Kai Chen, Yang Liu

Research Collection School Of Computing and Information Systems

Automatically fixing compilation errors can greatly raise the productivity of software development, by guiding the novice or AI programmers to write and debug code. Recently, learning-based program repair has gained extensive attention and became the state of-the-art in practice. But it still leaves plenty of space for improvement. In this paper, we propose an end-to-end solution TransRepair to locate the error lines and create the correct substitute for a C program simultaneously. Superior to the counterpart, our approach takes into account the context of erroneous code and diagnostic compilation feedback. Then we devise a Transformer-based neural network to learn the …

Go to article

Overview Of The Clpsych 2022 Shared Task: Capturing Moments Of Change In Longitudinal User Posts, Adam Tsakalidis, Jenny Chim, Iman Munire Bilal, Ayah Zirikly, Dana Atzil-Slonim, Federico Nanni, Philip Resnik, Manas Gaur, Kaushik Roy, Becky Inkster, Jeff Leintz, Maria Liakata Oct 2022

Overview Of The Clpsych 2022 Shared Task: Capturing Moments Of Change In Longitudinal User Posts, Adam Tsakalidis, Jenny Chim, Iman Munire Bilal, Ayah Zirikly, Dana Atzil-Slonim, Federico Nanni, Philip Resnik, Manas Gaur, Kaushik Roy, Becky Inkster, Jeff Leintz, Maria Liakata

Publications

We provide an overview of the CLPsych 2022 Shared Task, which focusses on the automatic identification of Moments of Change in longitudinal posts by individuals on social media and its connection with information regarding mental health . This year's task introduced the notion of longitudinal modelling of the text generated by an individual online over time, along with appropriate temporally sensitive evaluation metrics. The Shared Task consisted of two subtasks: (a) the main task of capturing changes in an individual's mood (drastic changes-`Switches'- and gradual changes -`Escalations'- on the basis of textual content shared online; and subsequently (b) the sub-task …

Go to article

Pandemic Time And Tourism In Oecd Countries: Artificial Intelligence And Digital Platforms, Alfonso Marino, Paolo Pariso, Michele Picariello Oct 2022

Pandemic Time And Tourism In Oecd Countries: Artificial Intelligence And Digital Platforms, Alfonso Marino, Paolo Pariso, Michele Picariello

University of South Florida (USF) M3 Publishing

Introduction underline the three phases related to sector crisis, Background, starting from literature highlight the importance of what are the main actions implemented in 38 Member States. Methodology, with SPAD, elaborates a qualitative and quantitative set of policy responses that are displayed in Results. Discussions highlight the different approaches within the OECD area, but also the absence of a common strategy to exit to the sector crisis. The conclusion emphasizes that crisis response policies still need to be built and developed in the OECD area, even though initial responses showed strong responses in individual Member States that did not address …

Go to article

Dynamic Temporal Filtering In Video Models, Fuchen Long, Zhaofan Qiu, Yingwei Pan, Ting Yao, Chong-Wah Ngo, Tao Mei Oct 2022

Dynamic Temporal Filtering In Video Models, Fuchen Long, Zhaofan Qiu, Yingwei Pan, Ting Yao, Chong-Wah Ngo, Tao Mei

Research Collection School Of Computing and Information Systems

Video temporal dynamics is conventionally modeled with 3D spatial-temporal kernel or its factorized version comprised of 2D spatial kernel and 1D temporal kernel. The modeling power, nevertheless, is limited by the fixed window size and static weights of a kernel along the temporal dimension. The pre-determined kernel size severely limits the temporal receptive fields and the fixed weights treat each spatial location across frames equally, resulting in sub-optimal solution for longrange temporal modeling in natural scenes. In this paper, we present a new recipe of temporal feature learning, namely Dynamic Temporal Filter (DTF), that novelly performs spatial-aware temporal modeling in …

Go to article

Machine Learning To Predict Warhead Fragmentation In-Flight Behavior From Static Data, Katharine Larsen Oct 2022

Machine Learning To Predict Warhead Fragmentation In-Flight Behavior From Static Data, Katharine Larsen

Doctoral Dissertations and Master's Theses

Accurate characterization of fragment fly-out properties from high-speed warhead detonations is essential for estimation of collateral damage and lethality for a given weapon. Real warhead dynamic detonation tests are rare, costly, and often unrealizable with current technology, leaving fragmentation experiments limited to static arena tests and numerical simulations. Stereoscopic imaging techniques can now provide static arena tests with time-dependent tracks of individual fragments, each with characteristics such as fragment IDs and their respective position vector. Simulation methods can account for the dynamic case but can exclude relevant dynamics experienced in real-life warhead detonations. This research leverages machine learning methodologies to …

Go to article

Hierarchical Hourglass Convolutional Network For Efficient Video Classification, Yi Tan, Yanbin Hao, Hao Zhang, Shuo Wang Oct 2022

Hierarchical Hourglass Convolutional Network For Efficient Video Classification, Yi Tan, Yanbin Hao, Hao Zhang, Shuo Wang

PhD Student’s Publications Collection

Videos naturally contain dynamic variation over the temporal axis, which will result in the same visual clues (e.g., semantics, objects) changing their scale, position, and perspective patterns between adjacent frames. A primary trend in video CNN is adopting spatial-2D convolution for spatial semantics and temporal-1D convolution for temporal dynamics. Though the direction achieves a favorable balance between efficiency and efficacy, it suffers from misalignment of visual clues with large displacements. Particularly, rigid temporal convolution would fail to capture correct motions when a specific target moves out of the reception field of temporal convolution between adjacent frames.To tackle large visual displacements …

Go to article

Two Singapore Public Healthcare Ai Applications For National Screening Programs And Other Examples, Andy Wee An Ta, Han Leong Goh, Christine Ang, Lian Yeow Koh, Ken Poon, Steven M. Miller Oct 2022

Two Singapore Public Healthcare Ai Applications For National Screening Programs And Other Examples, Andy Wee An Ta, Han Leong Goh, Christine Ang, Lian Yeow Koh, Ken Poon, Steven M. Miller

Research Collection School Of Computing and Information Systems

This article explains how two AI systems have been incorporated into the everyday operations of two Singapore public healthcare nation-wide screening programs. The first example is embedded within the setting of a national level population health screening program for diabetes related eye diseases, targeting the rapidly increasing number of adults in the country with diabetes. In the second example, the AI assisted screening is done shortly after a person is admitted to one of the public hospitals to identify which inpatients—especially which elderly patients with complex conditions—have a high risk of being readmitted as an inpatient multiple times in the …

Go to article

What Machines Can't Do (Yet) In Real Work Settings, Thomas H. Davenport, Steven M. Miller Oct 2022

What Machines Can't Do (Yet) In Real Work Settings, Thomas H. Davenport, Steven M. Miller

Research Collection School Of Computing and Information Systems

AI systems may perform well in the research lab or under highly controlled application settings, but they still needed human help in the types of real world work settings we researched for a new book, Working With AI: Real Stories of Human-Machine Collaboration. Human workers were very much in evidence across our 30 case studies. In this article, we use those examples to illustrate our list of AI-enabled activities that still require human assistance. These are activities where organizations need to continue to invest in human capital, and where practitioners can expect job continuity for the immediate future

Go to article

Multi-Functional Job Roles To Support Operations In A Multi-Faceted Jewel Enabled By Ai And Digital Transformation, Steven M. Miller Oct 2022

Multi-Functional Job Roles To Support Operations In A Multi-Faceted Jewel Enabled By Ai And Digital Transformation, Steven M. Miller

Research Collection School Of Computing and Information Systems

In this story, we highlight the way in which the use of AI enabled support systems, together with work process digital transformation and innovative approaches to job redesign, have combined to dramatically change the nature of the work of the front-line service staff who protect and support the facility and visitors at the world’s most iconic airport mall and lifestyle destination.

Go to article

Interactive Video Corpus Moment Retrieval Using Reinforcement Learning, Zhixin Ma, Chong-Wah Ngo Oct 2022

Interactive Video Corpus Moment Retrieval Using Reinforcement Learning, Zhixin Ma, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Known-item video search is effective with human-in-the-loop to interactively investigate the search result and refine the initial query. Nevertheless, when the first few pages of results are swamped with visually similar items, or the search target is hidden deep in the ranked list, finding the know-item target usually requires a long duration of browsing and result inspection. This paper tackles the problem by reinforcement learning, aiming to reach a search target within a few rounds of interaction by long-term learning from user feedbacks. Specifically, the system interactively plans for navigation path based on feedback and recommends a potential target that …

Go to article

Long-Term Leap Attention, Short-Term Periodic Shift For Video Classification, Hao Zhang, Lechao Cheng, Yanbin Hao, Chong-Wah Ngo Oct 2022

Long-Term Leap Attention, Short-Term Periodic Shift For Video Classification, Hao Zhang, Lechao Cheng, Yanbin Hao, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Video transformer naturally incurs a heavier computation burden than a static vision transformer, as the former processes �� times longer sequence than the latter under the current attention of quadratic complexity (�� 2�� 2 ). The existing works treat the temporal axis as a simple extension of spatial axes, focusing on shortening the spatio-temporal sequence by either generic pooling or local windowing without utilizing temporal redundancy. However, videos naturally contain redundant information between neighboring frames; thereby, we could potentially suppress attention on visually similar frames in a dilated manner. Based on this hypothesis, we propose the LAPS, a long-term “Leap …

Go to article

Wave-Vit: Unifying Wavelet And Transformers For Visual Representation Learning, Ting Yao, Yingwei Pan, Yehao Li, Chong-Wah Ngo, Tao Mei Oct 2022

Wave-Vit: Unifying Wavelet And Transformers For Visual Representation Learning, Ting Yao, Yingwei Pan, Yehao Li, Chong-Wah Ngo, Tao Mei

Research Collection School Of Computing and Information Systems

Multi-scale Vision Transformer (ViT) has emerged as a powerful backbone for computer vision tasks, while the self-attention computation in Transformer scales quadratically w.r.t. the input patch number. Thus, existing solutions commonly employ down-sampling operations (e.g., average pooling) over keys/values to dramatically reduce the computational cost. In this work, we argue that such over-aggressive down-sampling design is not invertible and inevitably causes information dropping especially for high-frequency components in objects (e.g., texture details). Motivated by the wavelet theory, we construct a new Wavelet Vision Transformer (Wave-ViT) that formulates the invertible down-sampling with wavelet transforms and self-attention learning in a unified way. …

Go to article

Towards Understanding The Faults Of Javascript-Based Deep Learning Systems, Lili Quan, Qianyu Guo, Xiaofei Xie, Sen Chen, Xiaohong Li, Yang Liu Oct 2022

Towards Understanding The Faults Of Javascript-Based Deep Learning Systems, Lili Quan, Qianyu Guo, Xiaofei Xie, Sen Chen, Xiaohong Li, Yang Liu

Research Collection School Of Computing and Information Systems

Quality assurance is of great importance for deep learning (DL) systems, especially when they are applied in safety-critical applications. While quality issues of native DL applications have been extensively analyzed, the issues of JavaScript-based DL applications have never been systematically studied. Compared with native DL applications, JavaScript-based DL applications can run on major browsers, making the platform- and device-independent. Specifically, the quality of JavaScript-based DL applications depends on the 3 parts: the application, the third-party DL library used and the underlying DL framework (e.g., TensorFlow.js), called JavaScript-based DL system. In this paper, we conduct the first empirical study on the …

Go to article

First Prev Next Last