About Me

🔎 About Me

I’m an M.Tech graduate in Data Science from IIT Madras (2024), and hold a B.Tech in Computer Science and Engineering from VIT (2022). I previously worked as a Project Associate at RBCDSAI, IIT Madras, under the guidance of Prof. B. Ravindran.

My research interests focus on reinforcement learning (RL), especially its applications to large language models (LLMs).

• 🔭 Current Focus: Reinforcement Learning for Large Language Models (RL for LLMs)
• 🤝 Open to Collaboration: Projects involving RL, LLMs, Causality, and Multi-Agent Systems
• 💬 Expertise Areas: Machine Learning · Deep Learning · Computer Vision · Natural Language Processing (NLP) · Reinforcement Learning

🚀 Research Interests

• Causality in Reinforcement Learning
• Reinforcement Learning
• RL for Large Language Models
• Multi-Agent Reinforcement Learning
• Multi-Agent Reasoning
• Representation Learning

🚀 Past Projects

Reinforcement Learning (RL)

• RL Assignments (IITM CS6700): Implementations including SARSA, Q-Learning, Hierarchical RL (SMDP & Intra-Option Q-Learning) , Dueling DQN, and REINFORCE.
  • Assignment 1 (SARSA/Q-Learning): Comparison of SARSA & Q-Learning in Grid World variants.
  • Assignment 2 (Dueling DQN/REINFORCE): Training Dueling DQN variants and Monte Carlo REINFORCE.
  • Assignment 3 (Hierarchical RL): Hierarchical RL in Taxi-v3 using SMDP & Intra-Option Q-Learning.
• RL4LLM Framework: Toolkit for RL-based Large Language Model fine-tuning.
• RL-Research Suite: Collection of benchmark Reinforcement Learning environments and algorithms.

Natural Language Processing (NLP)

• NLP Information Retrieval System (IITM CS6370): Project focusing on building an Information Retrieval system.
• Causal-NLP: Development of causal inference tools tailored for Natural Language Processing tasks.

Machine Learning (ML)

• Mathematical Essays on ML Algorithms: In-depth mathematical explorations of fundamental ML models.
  • Support Vector Machine: Mathematical essay on SVMs.
  • Random Forest: Mathematical essay on Random Forests.
  • Decision Trees: Mathematical essay on Decision Trees.
  • Naive Bayes Classifier: Mathematical essay on Naive Bayes.
  • Logistic Regression: Mathematical essay on Logistic Regression.
  • Linear Regression: Mathematical essay on Linear Regression.
• Advanced Regression - House Prices: Application of advanced regression techniques for house price prediction.

Deep Learning (DL)

• DL Assignment 1 (IITM CS6910): Implemented a feedforward neural network from scratch with multiple optimizers and hyperparameter tuning.
• DL Assignment 2 (IITM CS6910): Built a CNN from scratch, performed hyperparameter optimization, and applied interpretability techniques.
• DL Assignment 3 (IITM CS6910): Developed sequence-to-sequence models (RNN/LSTM/GRU) with attention for English-to-Malayalam transliteration.

Computer Vision (CV)

• BSc Thesis - COVID-19 CXR Classification: Classification of COVID-19 from Chest X-Rays using Transfer Learning.
• YOLO Object Detection: Projects utilizing YOLO models for object detection tasks.
  • Finding Cars from Aerial Images (YOLO-NAS/YOLOv8): Using YOLO-NAS and YOLOv8 to detect cars in aerial imagery.
  • Custom Dataset Object Detection/Segmentation/Classification (YOLOv8): Applying YOLOv8 for detection, segmentation, and classification on custom datasets.
• GreenVine - Early Plant Disease Detection: Computer vision project for early detection of plant diseases.
• Digit Recogniser: A computer vision project for recognizing handwritten digits.

🛠️ Languages & Tools

I have experience with a wide range of programming languages and tools, including:

• Programming Languages: Python, Java, C++, JavaScript, PHP, HTML5, CSS3
• Databases: MySQL, MongoDB
• Version Control & Big Data: Git, GitHub, Hadoop, Spark, Kafka
• Deep Learning & NLP: TensorFlow, PyTorch, Keras, Hugging Face
• Frameworks: jQuery, React, Node.js, Flask, ExpressJS, Bootstrap

📊 GitHub Stats

My GitHub profile showcases my active contributions to open-source projects and personal repositories focused on machine learning, reinforcement learning, and data science applications.

📬 Contact

Feel free to reach out at ahmecse@gmail.com or connect with me on LinkedIn!