Can dl4j achieve split learning?

Hi, I am making a project that need to use split learning. In the split learning, the forward propagation and the back-propagation take place in the following way:With the raw data, a client trains the network up to a certain layer of the network, called the cut layer, and sendsthe activations of the cut layer, also called smashed data, to the server. Then, the server carries out the training of the remaining layers with the smashed data that it received from the client. This completes a single forwardpropagation. Next, the server carries out the back-propagation up to the cut layer and sends the gradients of thes mashed data to the client. With the gradients, the client carries out its back-propagation on the remaining network(i.e., up to the first layer of the network). This completes a single pass of the back-propagation between a clientand the server.
I wonder to know can the deeplearning4j achieve this? if yes, which class or method should I refer to? thank you