This is a project based on attribute inference attack (or overlearning attack). We run the experiment on CelebA dataset.

Your proactive AI learning companion that turns daily work into continuous mastery.

WaniKani Userscript to automatically click the 'Need more time' button after lessons to enable overlearning

Collection of numerical experiments to do with the overlearning paper.

No description available

No description available

No description available

No description available

A simple 2D game made using Godot

A tool for overlearning the memory peg mnemonic system and getting response times to near instantaneous levels.

Using the tripadvisor dataset which is based on peoples traveling behaviour we aim at comparing key clustering algorithms with the aim of finding an optimal value that can be adopted in tourism domain and simultanously recommending most highly rated place to visit. The tripadvisor dataset is avalible on the public domain which we have used and applied the clustering techiniques line kmeans and Agglomerative. For this project we found LDA model giving a good accuracy 0.94 accuracy with precision being 0.97 and recall of 0.90. The randon forest was quite overfitting with 0.99 accuracy considering this being a small dataset and model overlearning. However PCA, Kernal-PCA did not improve the model performance here