top of page
City Skyline

Machine Learning Projects

Stay Home Safe with Starving Federated Data

Paper "Stay Home Safe with Starving Federated Data" has been accepted by the International Conference on Universal Village (IEEE UV2022). 

Proposed a novel robust federated adversarial training method named FLATS (Federated Learning Adversarial Training for Smart Home Face Recognition) under the supervision of Dr. Yajun Fang (CSAIL, MIT).
[paper] [code] [slides] [video]

robust_FL_diagram.png

General Architecture of FLATS

MSDT: Maksed Language Model Scoring Defense in Text Domain

Paper "MSDT: Masked Language Model Scoring Defense in Text Domain" has been accepted by the International Conference on Universal Village (IEEE UV2022). 

Proposed a novel improved backdoor defense method in text domain using Maksed Language Mode Scoring metric. "Independent Work Research (COMP4971D)" under the supervision of Professor, Minhao Cheng (HKUST)
[paper] [code] [slides] [video]

General Methodology of MSDT

Impact of Adversarial Training on the Robustness of Deep Neural Networks
 
Paper "Impact of Adversarial Training on the Robustness of Deep Neural Networks" has been accepted by the 2022 International Conference on Modeling, Simulation and Computing Science (MSCS 2022).

Experimented the effectiveness of various methods of  adversarial training on improving the robustness of neural networks against classifying perturbed histopathological images. 
[code]

source: "Explaining and Harnessing Adversarial Examples" paper

Histopathological Image Classification
 
Metastatic cancer diagnosis based on histopathological image using Convolutional Neural Network and modified Resnet-18.
[code] [slides] [report]

source: https://sergeykolchenko.medium.com/histopathologic-cancer-detection-as-image-classification-using-pytorch-557aab058449

Seq2Seq Neural Machine Translation (Keras)
 
Neural machine translation using Tensorflow framework. Translation processed from English to French
[code] [slides] 

Adversarial Attack and Defense Presentation Project
 
COMP4211 Machine Learning course Final Presentation Project on the topic of "Adversarial Attack and Defense".  [slides] [video]

 

Papers reviewed: 

  • “Explaining and Harnessing Adversarial Examples"

  • “Is Bert Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment”  

source: "Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment" paper

bottom of page