I'm

Akash Vartak

Graph Neural Networks , Deep Learning ,
Computer Vision , Natural Language Processing

Download CV

About

About Me

PhD Student and Research Assistant

I am pursuing my PhD in Computer Science @ University Of Maryland - Baltimore County in the Dept. of Computer Science and Electrical Engineering.

I work as a Research Assistant to Prof. Tim Oates in the CORAL Lab and work with Graph Neural Networks for deep learning model classification, Image Segmentation techniques, and Grounding Language using Visual Percepts.

I am open to collaborate on research projects or paper ideas. I am also open to internship opportunities as a ML/AI Intern or in a research position.
You can reach me at the email or phone number listed below.

Degree: PhD, Computer Science
Email: akashvartak.1995@gmail.com
University: University of Maryland Baltimore County
Phone: +1 (443)941-5919
Lab: Coral Lab, UMBC
Looking for: Internships
Hire Me

Qualifications

Qualifications

My Education

PhD in Computer Science

University of Maryland Baltimore County | 2022 - ONGOING

  • GPA: 3.89/4
  • Coursework: Artificial Intelligence, Natural Language Processing
  • Research: Graph Neural Networks, Adersarial ML (Backdoor Detection), Image Segmentation
Master of Science in Computer Science

University of Maryland Baltimore County | 2019 - 2021

  • GPA: 3.86/4
  • Coursework: Algorithms, Distributed Systems, Data Science, Machine Learning, Cryptography
  • Research: Using Text to Improve Classification of Man-Made Objects
Bachelor of Engineering in Computer Engineering

Savitribai Phule Pune University, formerly University of Pune | 2013 - 2017

  • GPA: 3.86/4
  • Coursework: Data Structures, Object Oriented Programming, Operating Systems, Artificial Intelligence

My Experience

Research Assistant

University of Maryland Baltimore County | MAY 2021 - ONGOING

  • Using graph Neural Networks for posisoned neural models.
  • Neural architecture search using deep learning ontologies.
Teaching Assistant

University of Maryland Baltimore County | AUG 2020 - MAY 2021

  • Interacting with students during and out of class and solving questions and queries.
  • Supervising graders in grading projects, assignments, homework, quizzes, and any other assessments.
Software Analyst

Yardi Software India Pvt. Ltd. | JUL 2017 - JUL 2019

  • Identified and mitigated application security issues which was verified by third-party security testing team.
  • Optimized invoice-batch processing times from 2 weeks to 1 hours, by implementing a smart, scalable scheduling algorithm.
  • Planned, designed and developed a complete end-to-end in-application chat function.
Project Intern

E2open | AUG 2016 - MAR 2017

  • Predicted best price of an item for a given store to maximize sales by developing a machine learning algorithm for sales forecasting and price optimization.

Skills

My Skills

  • Languages:
    Python, JAVA

  • Machine Learning/Data Science:
    PyTorch, PyTorch-Geometric, NetworkX, Scikit-learn, Pandas, NumPy, Matplotlib, Seaborn, BeautifulSoup, pyttsx, Shell scripting

  • Other Tools and Technologies:
    Git, LaTeX, Team Foundation Server (TFS), MS Office (Word, Excel, PowerPoint), WinSCP

  • Platforms:
    Linux, Windows

Research

My Research & Publications

Graph Neural Networks for Model Classification

  • Current research aiming to investigate the feasibility and performance of graph neural networks to identify and classify trojaned neural network models.

Using Text to Improve Classification of Man-Made Objects

A Survey on Promotional and Base Level Forecasting using ARIMA

  • Aakash Raina, Akash Vartak, Prof. P. R. Patil, Yash Katariya, Yash Lahoti, A Survey on Promotional and Base Level Forecasting using ARIMA, In International Journal of Computer Systems (IJCS), pp: 10-12, Volume 4, Issue 1, January 2017.
  • A survey on how the ARIMA forecasting model can be used in the retail industry for base level forecasts and for promotional sales prediction, specifically for Time Series data.

Projects

My Projects

Multi-Artist Recommendation System (MARS)

python, scikit-learn, matplotlib, seaborn, machine learning, data science

  • A system to recommend a list of artists to sing a song based on the lyrics, genre, mood, and other song and artist meta-data like the number of hits an artist has or the genre of an artist or tempo.
  • Recommendation model was based on Natural Language Processing and Naive-Bayes and SVM machine learning models.

Price Optimization and Elasticity

python, scikit-learn, matplotlib, seaborn, machine learning, data science

  • Prediction of an item's best price so as to maximize sales.
  • Developed a sales forecasting and price optimization ML model that achieved an 8-10% error rate on past sales data.
  • ML model used different regression models to predict the optimal price by analyzing past retail sales across a vast demographic data and of multiple store-item combinations.

Wordy

python, beautifulsoup, pyttsx

  • A personal vocabulary trainer that displays a word and its meaning as a desktop notification, originally made for Ubuntu OS.
  • Can also provide pronunciation using pyttsx library.

© 2022 Akash Vartak. All Rights Reserved.

Based on template by HTML Codex