A5017419487- Machine Learning and ELM
- CCD and CMOS Imaging Sensors
- Advanced Neural Network Applications
- Stochastic Gradient Optimization Techniques
- Privacy-Preserving Technologies in Data
- Human Mobility and Location-Based Analysis
- Neural Networks and Applications
Boğaziçi University
2024
Deep edge intelligence aims to deploy deep learning models that demand computationally expensive training in the network with limited computational power. Moreover, many applications require handling distributed data cannot be transferred a central server due privacy concerns. Decentralized methods, such as federated learning, offer solutions where are learned collectively by exchanging weights. However, they often complex devices may not handle and multiple rounds of communication achieve...
For image classification problems, various neural network models are commonly used due to their success in yielding high accuracies. Convolutional Neural Network (CNN) is one of the most frequently deep learning methods for applications. It may produce extraordinarily accurate results with regard its complexity. However, more complex model longer it takes train. In this paper, an acceleration design that uses power FPGA given a basic CNN which consists convolutional layer and fully connected...