Wearable sensor-based stress detection is a well-explored area of research in the domain of Affective computing and can be performed with the help of non-invasive sensing modalities like Electrodermal Activity (EDA). ...
详细信息
ISBN:
(纸本)9798350304367;9798350304374
Wearable sensor-based stress detection is a well-explored area of research in the domain of Affective computing and can be performed with the help of non-invasive sensing modalities like Electrodermal Activity (EDA). The EDA sensors with a wearable form factor are commonly available on commercial off-the-shelf devices. In recent years, with the increased availability of such wearable devices to end-users, these applications have become more pervasive and thus require a greater level of optimization for continuous usage on resource-constrained and battery-powered devices. While several research works have focused on designing Machine Learning and Deep Learning models for these tasks, very few focus on the resource footprint of these models. The balance between classification accuracy and computational resources is difficult to achieve with a manual design process and takes a long time. This is where automation plays a key role. In this work, we explore the applicability of automation Techniques such as Neural Architecture Search (NAS), to create tiny Stress Detection models suitable for round-the-clock inference, with minimal resource requirements, and low latency to perform in real-time, and show the trade-off between the various metrics on four publicly available datasets combined, and achieve an accuracy of 85.98% with a model size of 49.40 kB, which is comparable to the state-of-the-art counterparts.
The proceedings contain 54 papers. The topics discussed include: CHARM: adopting digitalization in community health assessment and review on mobile;comparative analysis of context based classification of twitter;augme...
ISBN:
(纸本)9781538671672
The proceedings contain 54 papers. The topics discussed include: CHARM: adopting digitalization in community health assessment and review on mobile;comparative analysis of context based classification of twitter;augmented reality technology in digital advertising;proposed urban travel behavior model to promote smart mobility;fake news on social media: brief review on detection techniques;learning analytics in Universiti Teknologi Brunei: Predicting Graduates Performance;The non-linearity effect of high power amplitude on OFDM signal and solution to solve by using PAPR reduction;middleware power saving scheme for mobile applications;and a hybrid trilateration and fingerprinting approach for indoor localization based on wifi.
This research study presents a comprehensive guide to MLOps best practices tailored for developing and deploying Artificial Intelligence (AI) based applications, focusing on the challenges and recent techniques in the...
详细信息
ISBN:
(纸本)9798350391558;9798350379990
This research study presents a comprehensive guide to MLOps best practices tailored for developing and deploying Artificial Intelligence (AI) based applications, focusing on the challenges and recent techniques in the field. Current systems face challenges such as maintaining model performance over time, ensuring scalability, platform automation, and handling complex deployment processes. Our proposed objective is to address these challenges through a detailed case study of an advanced Retrieval-Augmented Generation (RAG) based chatbot. This paper illustrates the critical components of an efficient Machine Learning Operations (MLOps) pipeline, covering the full life-cycle from data ingestion and model evaluation to continuous integration and deployment. Emphasis is placed on platform automation to streamline setup and operational processes, ensuring scalability, repeatability, and maintainability. Integrating MLOps practices aims to enhance the reliability and performance of AI applications, providing valuable insights and practical guidelines for practitioners in the field.
The global database automation market is booming due to the increasing data generated by various sectors. Consequently, organizations are aggressively pursuing solutions to improve operations and increase the efficien...
详细信息
In recent years, the Intent-driven Network (IBN) has been proposed to further enhance the intelligence of communication systems. In IBN, users can express their resource expectations through intents while the IBN perf...
详细信息
ISBN:
(纸本)9798350361261;9798350361278
In recent years, the Intent-driven Network (IBN) has been proposed to further enhance the intelligence of communication systems. In IBN, users can express their resource expectations through intents while the IBN performs resource scheduling to fulfill these intents. Many challenges of complex networks can be tackled by IBN such as the mobile edge computing (MEC) system. In a MEC system, limited computing resources are competed by the users which may cause the intent conflict. To solve this, in this paper, we introduce the IBN concept to the MEC system, where an intent conflict detection module is proposed. The proposed module is based on the Open Network automation Platform (ONAP) architecture. Moreover, by formulating the computing resource conflict problem as a Markov decision process (MDP) model, we employ an improved deep Qnetwork (DQN) algorithm to improve the efficiency of resource utilization. Simulation results demonstrate the completion time of intents is remarkably reduced in the proposed intent conflict resolution scheme.
Recently, an increasing demand for low-latency tasks such as VR/AR necessitates stringent latency requirements. Traditionally, low-latency tasks have been offloaded to base stations (BSs) with superior computing capab...
详细信息
The recognition of only visible image is difficult to meet the rapidly developing application requirements of increasingly complex application scenarios in automatic driving and industrial automation. More and more re...
详细信息
ISBN:
(纸本)9798400717840
The recognition of only visible image is difficult to meet the rapidly developing application requirements of increasingly complex application scenarios in automatic driving and industrial automation. More and more researchers have turned their attention to the field of multi-modal image fusion. In this paper, a stochastic fusion method of multi-modal image using infrared image as mask is introduced. Considering the deployment requirements of edge computing platforms, the recognition rate of fused image and non-fused image is compared and verified on several lightweight models. The results show that compared with non-fused images, the fusion image provided by this method can improve the recognition rate by 5%similar to 7% and reduce the storage cost of these fused images greatly(uint8 reduced to uint3). In a meanwhile, the accuracy of classification models for the stochastic VI-TIR fusion image is roughly equivalent to that of fusion images acquired by TarDAL and LDFusion methods. The results demonstrates that the stochastic fusion method is conducive to multi-modal image fusion and local recognition on the edge computing platforms.
The rapid advancements in edge computing have opened new avenues for optimizing real-time control systems in maritime environments. Traditional centralized systems face challenges such as high latency, limited scalabi...
详细信息
Imagine a home that adapts to your every need convenient, secure, comfortable, and efficient. This paper proposes a "smart mirror" solution, seamlessly blending a traditional mirror with advanced technology....
详细信息
Imagine a home that adapts to your every need convenient, secure, comfortable, and efficient. This paper proposes a "smart mirror" solution, seamlessly blending a traditional mirror with advanced technology. It displays personalized information within your reflection, from weather updates to calendar events, all controlled by voice commands. An integrated security system and essential item tracking add peace of mind. Testing confirms the model's effectiveness, paving the way for a smarter, more personalized living experience.
Point cloud prediction is an important yet challenging task in the field of autonomous driving. The goal is to predict future point cloud sequences that maintain object structures while accurately representing their t...
详细信息
ISBN:
(纸本)9798350384581;9798350384574
Point cloud prediction is an important yet challenging task in the field of autonomous driving. The goal is to predict future point cloud sequences that maintain object structures while accurately representing their temporal motion. These predicted point clouds help in other subsequent tasks like object trajectory estimation for collision avoidance or estimating locations with the least odometry drift. In this work, we present ATPPNet, a novel architecture that predicts future point cloud sequences given a sequence of previous time step point clouds obtained with LiDAR sensor. ATPPNet leverages Conv-LSTM along with channel-wise and spatial attention dually complemented by a 3D-CNN branch for extracting an enhanced spatio-temporal context to recover high quality fidel predictions of future point clouds. We conduct extensive experiments on publicly available datasets and report impressive performance outperforming the existing methods. We also conduct a thorough ablative study of the proposed architecture and provide an application study that highlights the potential of our model for tasks like odometry estimation.
暂无评论