Professor: Priyanka Golia

This SAT Solver is a sophisticated implementation of a Boolean satisfiability problem solver. It leverages several state-of-the-art techniques to efficiently determine the satisfiability of a given propositional logic formula in Conjunctive Normal Form (CNF). The solver adopts a hybrid strategy that combines various heuristic and algorithmic approaches for effective problem resolution.

Read more

Professor: Rohan Paul

This project is a collection of deep learning projects that I worked on. The first part of the project is a bird classifier, achieving an accuracy of nearly 93% on test set, and the other part is on representational learning on the MNIST dataset using a variational autoencoder (including GMM clustering on latent space).

Read more

Professor: Tarun Mangla

The Word-Counting Client project focuses on understanding socket programming, client-server communication, and the implementation of various scheduling algorithms in a networked environment. The project provides hands-on experience in designing a system where a client sends text data to a server, and the server processes the data to count the frequency of each word. The project incorporates both single-client and multi-client communication, along with the application of different scheduling strategies to manage how requests are handled by the server.

Read more

Professor: Rohan Paul

The goal of this project was to build a game-playing AI agent for the game Havannah. The core algorithm utilizes Monte Carlo Tree Search (MCTS) combined with Rapid Action Value Estimation (RAVE) and Frames (human intelligence) to efficiently explore and identify optimal moves in the game of Havannah.

Read more

Professor: Tarun Mangla

This project focuses on exploring the principles and practical implementation of Software-Defined Networking (SDN) using the OpenFlow protocol. SDN separates the control plane from the data plane, enabling centralized network management and dynamic policy implementation. The project involves hands-on development of key networking policies and functionalities to better understand the power and flexibility of SDN.

Read more

Professor: Tarun Mangla

The goal of this project is to design and implement a reliable file-transfer protocol using the User Datagram Protocol (UDP), which inherently lacks the reliability and congestion control mechanisms found in TCP. To address these limitations, the implementation adopts a hybrid approach that incorporates key principles from TCP to ensure robust and secure data transfer.

Read more

Professor: Rohan Paul

This project developed an Automatic Speech Recognition (ASR) corrector using local beam search to enhance ASR outputs by minimizing errors through phoneme mappings, word frequencies, and DP. The system uses a local beam search algorithm to explore multiple possible corrections for a given ASR output efficiently. By maintaining a focus on the most promising candidates during the search process, the algorithm balances computational efficiency and accuracy.

Read more

Professor: Sanjiva Prasad

ProCaml is a Prolog interpreter implemented in OCaml. Comprehensive documentation for ProCaml, including usage examples and technical details, is available to help users get started.

Read more

Professor: Kolin Paul

This project involves the implementation and testing of various cache replacement strategies, including Least Recently Used (LRU) and First-In-First-Out (FIFO). The simulation was conducted on large datasets to evaluate the efficiency and performance of each strategy in managing memory and reducing cache misses. The results provide insights into the effectiveness of different cache management techniques for optimizing computational tasks.

Read more

Professor: Kolin Paul

Developed a functional simulation of the RISC-V Instruction Set Architecture (ISA) using Logisim. Designed and implemented core components of the RISC-V processor, including the ALU, registers, and control unit

Read more

Professor: Huzur Saran

Implemented algorithmic trading strategies such as RSI, ADX, Mean Reverting and Pair Trading and Linear Regression. Evaluated strategies using historical data from the Indian Stock Market spanning the last 10 years

Read more

Professor: Rahul Garg

This program processes a given corpus in a specific format and answers user queries based on the information within the corpus. The goal is to develop a search mechanism that allows users to ask questions and retrieve relevant information from the corpus.

Read more

Professor: Preeti Ranjan Panda

Developed a neural network architecture implemented in VHDL to perform digit detection. The system reads images, processes them through the neural network, and displays the results on a VGA display

Read more

Professor: Huzur Saran

This project aims to diagnose individuals with Attention Deficit Hyperactivity Disorder (ADHD) through a series of interactive games.

Read more

Professor: Huzur Saran

This web application is developed as a part of the COP290 course. It's a Flask-based project designed to manage user registrations and logins, featuring a welcoming interface.

Read more

Professor: Subodh V. Sharma

Clutter Cutter is a versatile Python script designed to bring order to your file organization. This script takes advantage of Python's file handling capabilities to automatically sort files based on their names into respective directories.

Read more