roshanr11 / roshanr11

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Roshan Ram

linkedin


Contributions Welcome
visitor badge

I'm a final-year undergraduate student at Carnegie Mellon University, majoring in Information Systems with minors in Machine Learning and Software Engineering. I'm working towards making my mark on the world.

  • Machine Learning Engineer intern @  (2 x @ )
  • 🔭 Previously worked as a Machine Learning Engineer and Software Development Engineer intern with geospatial computer vision ML models @ In-Q-Tel, a venture capital firm that delivers cutting-edge technologies to the CIA and other US Government agencies
  • 🔭 Researching: multi-agent reinforcement learning, graph neural networks, zero-shot dependency parsing
  • 🌱 Frequent competitor on Kaggle
  • 🌱 Areas of Interest: Computer Vision, Natural Language Processing techniques, Deep Reinforcement Learning
  • 💬 Open to collaborate on or discuss Machine Learning papers/problems/projects
  • 📫 Reach me at: rram [at] andrew [dot] cmu [dot] edu

Projects:

General ML:

  • Using Neural Network to Create Bike-Sharing Predictions
  • Creating, Training, and Deploying a ML Model on AWS Sagemaker
  • Boston House Price Predictions
  • Interpreting Dermatalogist AI Results
  • Applying PCA, Unsupervised Learning to Learn Customer Segments from Data
  • Titanic EDA, Survival Predictions
  • Using XGBoost, Deep Learning to Predict Insurance Claims (Kaggle Competition, Capstone Project)

Computer Vision:

  • Using CNNs to Create a Dog-Breed Classifier
  • Face Generation Using Generative Adversarial Networks
  • Facial Keypoint Detection
  • Image Captioning with CNNs + RNN/LSTM on MS-COCO Dataset
  • Landmark Detection & Robot Tracking (SLAM)

Natural Language Processing:

  • Creating TV Captions/Scripts with RNNs
  • Deep Neural Networks for Speech Recognition
  • Hidden Markov Models for Part-of-Speech (POS) Tagging
  • RNNs for Machine Translation
  • RNNs for Text Sentiment Analysis

Reinforcement Learning:

  • Applying RL to Solve Frozen-Lake Environment Problem
  • Applying RL to Create A BlackJack AI
  • Applying RL to Solve Cliff-Walking Environment Problem
  • Applying RL to Solve OpenAI's Taxi-World Environment Problem
  • Applying RL to Deep Q-Learning
  • Applying RL, Deep Q-Learning to Teach A Quadcopter How To Fly

Other:

About

License:MIT License