Unofficial JAX implementations of deep learning research papers

We implement RCNN algorithm for object detection from an Images.

This repository is accompanying the papers "Artificial Fast Fading from Reconfigurable Surfaces Enables Ultra-Reliable Communications" (SPAWC 2021), "RIS-Assisted Statistical Channel Shaping for Ultra-High Reliability" (WSA 2021), and "Reconfigurable Intelligent Surface Phase Hopping for Ultra-Reliable Communications" (TWC, DOI:10.1109/TWC.2022.3172760).

Using NeRF we can convert 2d images into 3d.

Research paper Implementation: Natural-based underwater image color enhancement through fusion of swarm-intelligence algorithm

Pytorch Implementation of Capsule Networks

Using Tapestry algorithm in Elixir to construct a fault-tolerant distributed hash table for sharing files in a p2p network

Minimum Hitting Set solver described in my master thesis and a subsequent research paper

A collection of research paper implementations in PyTorch

This repository belongs to the paper "Reliability and Latency Analysis for Wireless Communication Systems with a Secret-Key Budget" (K.-L. Besser, R. F. Schaefer, H. V. Poor, IEEE Transactions on Communications, vol. 72, no. 2, pp. 1033-1044, Feb. 2024).

My implementation(s) of ChessCrypt, focusing on the core S-Box generation algorithm using chess piece movements for cryptographic strength

A secure elliptic curve cryptography based mutual authentication protocol for cloud-assisted TMIS

Implementation of semi and self supervised learning on Imbalanced Dataset

This repository is accompanying the paper "Multi-User Frequency Assignment for Ultra-Reliable mmWave Two-Ray Channels" (WiOpt 2022, DOI:10.23919/WiOpt56218.2022.9930571)

This repository contains the implementation of the Self Meta Pseudo Labels (SMPL) method for semi-supervised learning

This code is the implementation of the research paper: New insights on integer programming models for the kidney exchange problem. Miguel Constantino, Xenia Klimentova, Ana Viana, and Abdur Rais (Constantino et al). European Journal of Operational Research, 231(1):57–68, 2013. Integer programs in the paper are solved using glpk solver in python.

My implementation of Machine Learning and Deep Learning papers from scratch.

Apply Style of one image onto another image using VGG19 Deep Neural Network in PyTorch

A reimplementation of the paper Human-Aligned Image Models Improve Visual Decoding from the Brain

Github repository for the "Foveated Animations for Efficient Crowd Simulation" project.

We proposed an approach that use the keywords of research paper as feature and generate a Restricted Boltzmann Machine (RBM).

This repository is accompanying the paper "Frequency Diversity for Ultra-Reliable and Secure Communications in Sub-THz Two-Ray Scenarios" (Karl-L. Besser, Eduard Jorswieck, and Justin Coon, IEEE ICC 2023. doi:10.1109/ICC45041.2023.10279098).

This project develops a novel agentic AI architecture for detecting and classifying adaptive and maladaptive self-states in mental health posts from social media, specifically Reddit. Unlike traditional approaches that rely solely on large language models (LLMs) for direct classification, this system uses modular reasoning-action agents, retrieval.

[IEEE AIIoT 2025] The official code for the paper "A Distributed Edge Computing Prototype Using Raspberry PIs"

FrugalAI Chip: Modular silicon architecture for disposable AI. Achieves 10.9× better CAPEX efficiency using 28nm chiplets + algorithmic decomposition. Validated: +4.8% accuracy, -91% carbon, 21.47× transformer speedup.

A from-scratch PyTorch implementation and tutorial of the landmark 2003 Neural Probabilistic Language Model paper.

This project is the implementation of the research paper titled 'Deep Convolutional Neural Networks for Sentiment Analysis of Short Texts'

This code is the implementation of the research paper: Statistical change detection for multi-dimensional data. Xiuyao Song, Mingxi Wu, Christopher Jermaine, and Sanjay Ranka. 2007 In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining. Association for Computing Machinery,New York,NY,USA,667–676.

This project implements an Automatic Essay Scoring (AES) system that combines Machine Learning with Ontology-based evaluation for intelligent essay assessment. Developed as part of the Knowledge Representation & Reasoning (KRR) course, it demonstrates the practical application of semantic knowledge structures in educational technology.

My implementation of the paper "Feature engineering strategies for credit card fraud detection"