Bài giảng "Trí tuệ nhân tạo" Chương 6: Perceptron, giới thiệu mô hình mạng nơ-ron đơn giản đầu tiên dùng trong học máy. Nội dung bài học giúp hiểu cách hoạt động của Perceptron, thuật toán huấn luyện, và khả năng phân loại tuyến tính. Đây là nền tảng để tiếp cận mạng nơ-ron và deep learning.

Efficient water resource management is a critical mandate for governmental authorities, as it directly impacts the effective utilization of this invaluable natural resource. The expeditious and accurate extraction of water surfaces significantly impacts governmental decision-making.

This study presents a novel polyp segmentation approach using ResUnet++. Trained on Kvasir-SEG and CVCClinicDB, ResUnet++ significantly outperforms traditional UNet and ResUnet. Its residual blocks and attention mechanisms enhance feature extraction, leading to improved segmentation in challenging cases.

In this article, we will discuss the use of a database, called Churn Modeling, which collects statistical data from banks. We will also explore the application of the BernoulliNB algorithm, combined with the incremental machine learning method, to process streaming data and analyze and predict customer churn rates in banks.

The present study aims to explore the utilization of a depression symptom database, employing classical machine learning techniques, with a focus on the Random Forest algorithm alongside other methodologies, to assess and diagnose stress levels.

The research focuses on applying deep learning to automate the fruit recognition and classification process, meeting the development needs of modern agriculture. Applying this technology helps improve efficiency and classification quality and reduces labor costs, resulting in lower product prices. The research team used two deep learning models, SSD300 and YOLOv10s, to recognize and classify six types of fruits: apples, bananas, kiwis, lemons, oranges, and strawberries.

Artificial intelligence is gradually emerging as a method for optimizing various tasks, offering cost-saving and highly efficient solutions. Nowadays, Al is used as a general term for diverse tasks performed by computers. Fields like machine learning, deep learning, and data science, among others within this scope, are considered part of Al as long as they exhibit the characteristics of artificial intelligence.

Bài giảng "Máy học nâng cao: Deep learning - An introduction" cung cấp cho người đọc các nội dung: Introduction, applications, convolutional neural networks vs. recurrent neural networks, hardware and software. Mời các bạn cùng tham khảo nội dung chi tiết.

Automatically identifying and understanding students' questions about problems they encounter or issues related to their universities is very important for universities to promptly grasp the aspirations of their students. This enables them to support and satisfy their students and enhance their reputation. Especially as social networks and online media continue to develop, students can easily post their questions and concerns online.

This paper proposes an intelligent plastic waste detection and classification system based on the Deep Learning model and Delta robot. This system includes a Delta robot, a camera, a conveyor, a control cabinet, and a personal computer.

This paper also discusses practical applications of object detection in satellite images, including environmental monitoring, resource management, and disaster response. Finally, the paper suggests potential future research directions, such as developing more efficient models, handling small objects, and leveraging diverse data sources.

This paper presents a lightweight deep learning-based product object classification scheme designed for deployment on edge servers. Leveraging the ImageNet Large Scale Visual Recognition Challenge 2012 (ILSVRC2012) dataset, six classes relevant to product objects are selected for model training and evaluation.

This paper comprehensively explores and compares methods for multi-person action recognition, with a focus on integrating YOLOv7-Pose a tool known for its rapid pose estimation capabilities--with deep learning architectures. Specifically, it examines the use of Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) and Spatial TemporalGraph Convolution Network (ST-GCN) to achieve precise action classification.

Severe air pollution in Vietnam's tourism areas has become a significant economic issue in recent years. While many studies have found a link between population exposure to air pollution and poor health outcomes, short-term exposure to air pollutants in high-pollution zones can result in acute health consequences; thus, poor air quality jeopardizes visitors' health and well-being and threatens the tourism industry's sustainability.

In this paper, we propose a novel facial wrinkle segmentation method based on the Unet++ model, enhanced with dice and focal loss functions. Our approach begins with the construction of an enriched wrinkle dataset sourced from the Flickr-Faces-HQ dataset, ensuring diversity in wrinkle types and complexities.

Luận văn "Xây dựng hệ thống nhận dạng kiểm soát khuôn mặt với Deep Learning" nghiên cứu kỹ thuật phát hiện ảnh mặt dựa trên đặc trưng HOG (Histogram of Oriented Gradients); nghiên cứu thuật toán giám sát SVM (Support Vector Machine); nghiên cứu thư viện mã nguồn mở OpenCV; nghiên cứu thư viện mã nguồn mở Dlib  Nghiên cứu thư viện mã nguồn mở Tensorflow và Keras; nghiên cứu kiến trúc mạng CNN VGG16; nghiên cứu áp dụng hàm Triplet Loss của mạng Deep Learning cho bài toán nhận dạng mặt.

This paper presents initial experiments on using deep learning to identify pulmonary diseases through X-ray image recognition. In experiments, there were three pulmonary diseases: aortic enlargement, lung opacity, and another lesion.

Blur image classification is essential for computer vision applications, including image quality assessment, surveillance and medical imaging systems. This study proposes a method to classify different types of blur: sharp, Gaussian blur, motion blur, and defocus blur, using the DenseNet121 architecture.

Bài viết này đề xuất một tiếp cận phát hiện và phân loại một số bệnh trên trái cam với tập dữ liệu được lấy từ Kaggle kết hợp sử dụng một số kỹ thuật học sâu và thị giác máy tính dựa trên kiến trúc mạng YoloV8 và mạng MobileNetV2.