Comparative Study of Convolutional Neural Network with other Machine Learning Algorithms
As evident from the project title, we will undergo a comparative analysis of Machine Learning (ML) and Deep Learning (DL) algorithms. In the first phase, we will focus on the underlying details of various machine learning algorithms and analyze why machine learning algorithms stuck at point of zero
2025-06-28 16:30:52 - Adil Khan
Comparative Study of Convolutional Neural Network with other Machine Learning Algorithms
Project Area of Specialization Artificial IntelligenceProject SummaryAs evident from the project title, we will undergo a comparative analysis of Machine Learning (ML) and Deep Learning (DL) algorithms. In the first phase, we will focus on the underlying details of various machine learning algorithms and analyze why machine learning algorithms stuck at point of zero improvement even after increasing the amount of training data. We will look for reasons as to why State Of The Art (SOTA) Machine Learning algorithms are less accurate than a generic Deep Learning algorithm in a certain problem space. Moreover, we will study in detail the anatomy of DL algorithms and look for the reasons how to tune the Deep Learning algorithms to generalize well. We will also carry out in depth analysis of what are the problem areas where machine learning algorithms are established as gold standard. Furthermore, we will also explore the problem areas where deep learning is preferred over classical machine learning algorithms given that increasing the amount of data.
Project ObjectivesOur objective is the comparative analysis of of Deep learning and Machine learning models. Our research will focus on the demystyfication of inner the workings of various algorithms and ways to improve their performance. Moreover, we will be researching how and why machine learning algorithms work better than deep learning for specific problem and vice versa.
Project Implementation MethodOur project is based on Software and Hardware. The algorithum of machine learning and Deep learning require high computation which require GPU and RAM. Several tools including PyCharm, Matlab.
Benefits of the ProjectOur research will help people in optimization of existing machine and deep learning algorithms for a specific type of data. Moreover, a comprehensive analysis of the computational power requried for a dataset will also be studied which will help other people in the implementation of machine and deep learning models.
Technical Details of Final DeliverableOur research based project will deliver a brief overview of the strength and weaknesses of existing machine and deep learning models. Moreover, a brief analysis of the algorithms will be provided in terms of their hyper parameters. In addition, code suppliments will also be included for future use in research work promoting the slogan of papers with code.
Final Deliverable of the Project Software SystemCore Industry ITOther Industries Education Core Technology Artificial Intelligence(AI)Other Technologies Artificial Intelligence(AI)Sustainable Development Goals Quality Education, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 79000 | |||
| GPU NVIDIA GEFORCE RTX 2080 SUPER | Equipment | 1 | 39000 | 39000 |
| Thesis | Miscellaneous | 3 | 1000 | 3000 |
| Book Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow 2 | Miscellaneous | 1 | 7000 | 7000 |
| SSD 512 GB+ Internet | Equipment | 2 | 15000 | 30000 |