Available Projects

Published in Digital Science, 2021

This post covers projects available for FYP/Research students. Students may also contact with their project proposals.

Adversarial Machine Learning

  • The objective of the project is to study different vulnerabilities in Deep Learning Algorithms and propose a solution or analyze recently published solutions to mitigate the attacks. The project study includes in particular the vulnerabilities of Convolutional Neural Network in the image classification that has direct applications in facial recognitions technology, health applications, and self-driving or autonomous vehicles.
  • Skills: Python, PyTorch, OpenCV

Melanoma Classification

  • The objective of the project is to train a machine learning model that can identify skin cancer in particular Melanoma. The project also includes deploying the model as a web app.

  • Dataset: https://www.kaggle.com/c/siim-isic-melanoma-classification/data [Any other similar dataset may be used for similar medical imaging task]

  • Skills: Python, PyTorch, OpenCV

Conversational AI - Chatbot

  • Build a chat bot using dataset that has conversational exchanges between pairs and deploy the model online
  • Dataset: https://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html [Any other dataset can also be used or a new dataset may be created]
  • Skills: Python, PyTorch, OpenCV

Data-Centric AI

  • https://https-deeplearning-ai.github.io/data-centric-comp/
  • In most machine learning projects, we build algorithm and tune hyper-parameters for fixed dataset. In the data-centric AI approach, we fix the model but improve the dataset. The improving dataset may include fixing incorrect labels, adding examples that represent edge cases, apply data augmentation, etc.
  • Skills: Python, PyTorch/Tensorflow, OpenCV