Lin Li's repositories
DA-Alone-Improves-AT
data augmentation alone can improve adversarial training
Combating-RO-AdvLC
Combating robust overfitting in adversarial training via AdvLC
Instance-adaptive-Smoothness-Enhanced-AT
Instance adaptive Smoothness Enhanced Adversarial Training (ISEAT)
Facial-Expression-Recognition_Decision-Tree
Facial Expression Classificaion using Decision Trees with input of AUs
adversarial-robustness-toolbox
Python library for adversarial machine learning (evasion, extraction, poisoning, verification, certification) with attacks and defences for neural networks, logistic regression, decision trees, SVM, gradient boosted trees, Gaussian processes and more with multiple framework support
cifar10-r
CIAFR10-R(endition): a downsampled variant of ImageNet-R(endition)
CNTK-Hotel-pictures-classificator
This project is a part of my MSc degree in Department of Computing, Imperial College London. This project is done under the efficient collaboration with my classmates Tomasz Bartkowiak, Danlin Peng, Yini Fang, Suampa Ketpreechasawat, and Nattapat Chaimanowong. We appreciate much all sincere and helpful advices received from co-supervisors, Anandha Gopalan (Imperial College London), Lee Stott (Microsoft), Tempest van Schaik (Microsoft) and Geoff Hughes (Microsoft), and the sponsor from Microsoft regarding Azure.
inner-detectors
An inner detector is a unit, representing a semantic concept in the raw input image, in an intermediate layer. This project is designed to identify inner detectors, explore their properties and transfer them. The goal of project is to build an high interpretable network via transferring inner detectors.
interpreting-techniques
the implementation of selected popular techniques for interpreting the complicated machine learning models esp. Deep Neural Networks.
OODRobustBench
The code base of OODRobustBench
RNN-from-scratch
This is a tutorial for personal practice regarding building RNNs from scratch.
treelli.github.io
A beautiful, simple, clean, and responsive Jekyll theme for academics