vaibhavbh-0 / cs418-final-project Goto Github PK

I am a master's degree graduate from UIC in Computer Science. My interests are in Data Science and in application of Machine Learning on structured and non-structured data such as images, videos and text.

I am a candidate with demonstrated strong analytical, critical thinking and communication skills with 3 years of work experience in Mobile App Development and Machine Learning. And an additional 6 months of professional & coursework experience in machine learning, deep learning, data science, natural language processing and computer vision.

Following are the list of my completed and ongoing relevant projects:

• Developed novel ways to improve disentangled representations by estimating Total Correlation across stochastic layers of Hierarchical VAE, NVAE.
• Implemented a multi-modal text and image model for SemEval 2023 Task on Google Cloud Platform inspired by CLIP and LiT. Retrieved relevant image conditioned on text context with a custom contrastive loss function based on this task.
• Constructed a data pipeline for YOLO v7 for real-time object detection with mosaic data augmentation.
• Built a classification and outlier detection model for cyber-attacks. Performed data preprocessing, EDA on the dataset with more than 16 million records. and 83 statistical features with. Led a team of 3 students/peers on AWS Sagemaker.
• Built a chatbot application for healthcare which assessed the user’s health and psychological correlates from the user’s interaction and provided a summary of the stylistic analysis of the user.
• Built a translation model based on Transformer architecture, which translates programming language to natural language. The model was trained on a TPU Node in GCP.
• Implementation of Spatial Transformer Network, it helps in making the convolution neural network robust to affine transformations.

To learn more about me check out my LinkedIn profile.