BEng thesis project on Binarized Neural Networks architectures.
-
Updated
Aug 7, 2023 - Jupyter Notebook
BEng thesis project on Binarized Neural Networks architectures.
A structured ensemble of Binarized Neural Networks for image classification on Raspberry Pi Pico
This repository contains the python code related to the Master's Thesis. In particular, this code is in charge of setting up the NN and training them, calculating the importance and developing the design procedures for reducing the final logic implementation complexity.
Binary classification between x-ray images with covid or pneumonia
Dimensionality Reduction in Images for Neural Networks.
A simple deep neural net class written to work with Numpy and Cupy
This project involves training the game "Misere Tic Tac Toe" to the computer using Robust and Efficient Reinforcement Learning Techniques
BNN trained on Kannada MNIST
BNN verification dataset for Max-SAT Evaluation 2020
Progressive Neural Architecture Search coupled with Binarized CNNs to search for resource efficient and accurate architectures.
My PhD thesis with all its source files, including all .tex files and images created, as well as the slides of my defense.
The official repository for the paper LAB: Learnable Activation Binarizer for Binary Neural Networks.
Introduce a Software Acceleration filter for Real Time Object Recognition with Binary Neural Network on FPGA
Code implementation of our AAAI'22 paper "Improved Gradient-Based Adversarial Attacks for Quantized Networks"
An implementation of the Binarized Neural Networks
Binarized Neural Network DoA estimation
Code implementation of our AISTATS'21 paper "Mirror Descent View for Neural Network Quantization"
Neural Property Approximate Quantifier
Deep Learning Framework with a specialisation aimed for Binarized Neural Networks.
Add a description, image, and links to the binarized-neural-networks topic page so that developers can more easily learn about it.
To associate your repository with the binarized-neural-networks topic, visit your repo's landing page and select "manage topics."