Because of the intellectual property protection laws are so strict in Germany, downloading with p2p tools like baidunetdisk or xunlei online is very dangourous and is likely to be discovered by online police at any time. So we can try to replace these p2p tools with Aria2 that is faster and sampler.
6-Link-Prediction-for-Knowledge-Graphs
5-Recent-Developments-in-Robustness-against-Adversarial-Perturbation
This project sums up the main points of the recent developments in deep learning for robustness against adversarial perturbation, then elaborates the tasks, datasets, evaluations of the previous approach adversarial training and the state of the art parseval regularization. Furthermore, a detailed comparison was made between of the previous methods and the most advanced approaches. A summary is provided at the end of this work, aiming to provide more new ideas and directions for future work.
4-mlrPlayground
Interactive machine learning playground for the mlr package done in shiny.
3-p2p-matrix-Rust-library-for-distributed-matrix-multiplication
2-landmARTigo-A-game-that-allows-user-to-identify-and-geolocate-landmarks-depicted-in-artwork
This project draft envisions the human computation system landmARTigo that leverages the human perception and knowledge of real users to identify landmarks that might
be depicted in artworks. The system is similar to the ESP game, ARTigo and Karido. To incentivize users, we elaborated several benefits for users and stakeholders. We also developed a concept for the user interface. The draft includes a model for data processing and the algorithms involved in the human computation system. The system has to deal with the cold-start problem, which we want to mitigate with several solutions. We devised metrics for evaluation and how to execute the evaluation. For future work, there are several extensions we came up with, that are possible, if the system is successful.