| KEYWORD |
Deep Learning Architecture Designer System Using Visual Blocks for User-Friendly Model Creation
keywords ARTIFICIAL INTELLIGENCE, DEEP LEARNING, DEVELOPERS, HUMAN COMPUTER INTERACTION, MACHINE LEARNING, MODEL, USER INTERFACE, VISUAL PROGRAMMING
Reference persons LUIGI DE RUSSIS
External reference persons Tommaso Calò
Research Groups DAUIN - GR-10 - Intelligent and Interactive Systems - e-LITE
Thesis type EXPERIMENTAL, RESEARCH
Description Deep learning has revolutionized numerous areas of research and application, with architectures of varying complexity being designed for different tasks. However, creating these architectures often requires expertise in specific libraries and a deep understanding of neural networks, making it inaccessible to many. The advent of visual programming and block-based interfaces provides an opportunity to bridge this gap.
This thesis aims to design and implement of a deep learning architecture designer system that uses visual blocks, making the process of building machine learning models more accessible and intuitive, especially for those without a deep technical background.
The main goals of this thesis are:
1. Review and understand the current state of the art in deep learning architectures and visual block-based design interfaces.
2. Design and implement a deep learning architecture designer system using visual blocks.
3. Conduct user testing with a small pool of users to evaluate the system's usability and efficiency.
If satisfactory, the result of the thesis will be released as an open-source project.
Required skills Knowledge in deep learning frameworks (e.g., TensorFlow, PyTorch) is beneficial.
Experience with UI/UX design will be advantageous.
Deadline 11/03/2024
PROPONI LA TUA CANDIDATURA