Network embedding, also known as network representation learning, aims to represent the nodes in a network as low-dimensional, real-valued, dense vectors, so that the resulting vectors can be represented and inferred ...
详细信息
Within the field of games, visualization of player log data is becoming an important method for its utility in providing an intuitive and informative way to understand players' experience, which is thus often used...
Within the field of games, visualization of player log data is becoming an important method for its utility in providing an intuitive and informative way to understand players' experience, which is thus often used by game analytics personnel and game user researchers. Moreover, even players themselves show increased interest in using analytics to quantify and self-improve their performances. Among other types of visualizations, node-edge graphs have proven to be capable of revealing the process of individual and aggregated players, allowing analysts to discover play patterns that can inform game design. However, visualization of player traces often has several disadvantages. First, displaying players' process data does not trivially scale, as tendentially high variance often leads to complex and abstruse graphs. Second, when aggregating all players, individual variations are often overlooked. For example, data from minorities (e.g., casual players or players who played the game very differently than others) are often treated as outliers or noise. In this paper, we present an iterative segmentation approach that allows analysts to interact with the visualization and group players into different subcategories through meta-data or behavioral patterns. Using this approach, analysts can bypass complicated visualizations while protecting significant unique information.
This paper presents a tool for monitoring and visualization of Robot Operating System (ROS) data in a standard browser. It is compatible with ROS and uses ReactJS to visualize topic data. The browser-based solution is...
详细信息
ISBN:
(纸本)9781728185040
This paper presents a tool for monitoring and visualization of Robot Operating System (ROS) data in a standard browser. It is compatible with ROS and uses ReactJS to visualize topic data. The browser-based solution is software-agnostic and does not require ROS installation to view data, which are sent via ROS topics. It is compatible with most browsers and could be deployed on a remote hosting. The deployed solution could be accessed by a remote user with no need to connect the robot's local network. The developed software was tested on Turtlebot 3 robot in the Gazebo simulator. It has a modular structure and could be integrated into any robotic system based on ROS.
This paper introduces an algorithm for autonomous self-modeling of robots through the integration of Large Vision Model (LVM) and Large Language Model (LLM). Our approach differs from traditional robotic approaches in...
详细信息
ISBN:
(数字)9798350364194
ISBN:
(纸本)9798350364200
This paper introduces an algorithm for autonomous self-modeling of robots through the integration of Large Vision Model (LVM) and Large Language Model (LLM). Our approach differs from traditional robotic approaches in that it enables robots to independently discover and refine their own body structure and control strategies using only partial information. Through a symbiotic process that includes LLM’s ability to generate predictive control code based on finite prompts, and LVM’s visual reasoning to validate and improve those predictions, our algorithm facilitates a self-learning loop. This cycle is characterized by an inner loop of assumptions, observations, and adjustments, supplemented by an outer loop that gradually increases the information provided until convergence is reached. The effectiveness of the process was quantified by measuring the difference between the expected and actual joint motion as a cost function to determine the minimum feasible prompt (MVP). Simulation results indicate that the algorithm is capable of self-modeling of with minimal initial information.
The frame-based method is not suitable for counting repetitive actions in event stream, since the framing process will disrupt the temporal information of events. For accurate count of repetitive actions in events, we...
详细信息
ISBN:
(数字)9798350349399
ISBN:
(纸本)9798350349405
The frame-based method is not suitable for counting repetitive actions in event stream, since the framing process will disrupt the temporal information of events. For accurate count of repetitive actions in events, we propose a framework based on threefold ideas: a) converting event stream into time series, b) searching candidates of repetitive actions based on the ascending and descending trends of event time series, and c) checking the candidates with a fast dynamic time warping based method. For accurate counting repetitive actions, an action enhancement method for event time series and a Mann-Kendall test incorporated dynamic candidate selection algorithm are innovatively proposed. The experimental results on artificially synthesized and normally recorded event datasets demonstrate that our framework can count repetitive actions in event stream with high accuracy. All codes, datasets and examples of visualization can be found at https://***/ZYL618/action_count_in_events3.
In addition to sampling the ocean by conventional platforms such as research vessels, a wide variety of instruments and autonomous vehicles, today's marine research increasingly relies on measurements by private i...
详细信息
Much research has been done on the detection and tracking of paved, preferably marked roads. Less work is available on the detection of dirt roads. The challenge is to provide a framework to track both paved roads and...
详细信息
ISBN:
(数字)9781737749721
ISBN:
(纸本)9781665489416
Much research has been done on the detection and tracking of paved, preferably marked roads. Less work is available on the detection of dirt roads. The challenge is to provide a framework to track both paved roads and dirt roads. In this paper, we are addressing the problem of developing measurement approaches working for both kinds of roads likewise. For that we fuse LiDAR with vision: First, we present indirect measurements from a static environment model populated with LiDAR data, as well as a new approach for LiDAR measurements from a segmented point cloud. Second, we investigate different image color modes to improve the effectiveness of locating dirt road boundaries using local oriented edge detection. We demonstrate the robustness of our measurements on difficult roads by showing qualitative results from our autonomous vehicles.
Malware detection and classification systems are essential components of cybersecurity. The identification and categorization of various unknown and variant malware strains present an ongoing challenge. Current litera...
详细信息
ISBN:
(数字)9798350377002
ISBN:
(纸本)9798350377019
Malware detection and classification systems are essential components of cybersecurity. The identification and categorization of various unknown and variant malware strains present an ongoing challenge. Current literature indicates that visualization-based malware detection and classification models leveraging deep learning demand substantial datasets, high computational resources, significant storage space, and extensive training times. There is a pressing need to reduce these computational costs, storage requirements, and training durations while enhancing model performance to achieve greater efficiency. This paper contributes by proposing an improved malware classification model that employs 1D-CNN and stacked ensemble techniques. In our experiments, we utilized two sets of custom and classical 1D-CNN models as weak learners. The top-performing weak learners were selected to create two ensemble models. Among these, the best-performing ensemble model achieved a classification accuracy of 98.01% with minimal computational resources, significantly improving efficiency.
The Safe Hydrogen Agile Pipeline Engineering (SHAPE) project aims to develop and study an innovative Digital Twin (DT) system enhanced by eXtended Reality (XR) and Artificial Intelligence (AI) technologies to improve ...
详细信息
visualization of microvasculature in Optical Coherence Angiography (OCA) is based on discrimination of red blood cells motion against the surrounding motionless tissue. In practice, surrounding living tissue is never ...
详细信息
暂无评论