Zeppelin as generic notebook interpreter system for cluster computing is capable of connecting different distributed systems. The main focus right now is Apache Spark which itself have extensive machine learning facilities.
There are other systems such as H2O, PredictIO, Parameterserver or Singa implementing advanced techniques based on neural networks, commonly referred as Deeplearning.
The goal of this Google Summer of Code project is to develop a tutorial Notebook with a sample data and extend Zeppelin server to be able to show advantages of interactive machine learning.
Stunents are welcome to reach out Zeppelin community via its developers mailing list to discuss the details, see http://zeppelin.incubator.apache.org/community.html