We present gradient routing, a way of controlling where learning happens in neural networks. Gradient routing applies masks to limit the flow of gradients during backpropagation. By supplying different masks for different data points, the user can induce specialized subcomponents within a model. We think gradient routing has the potential...
I'm a student in a subfield of CS and biology, getting my bachelor's degree later this year and planning on continuing to masters (and maybe beyond). My ultimate goal is to work on the edge between biology, CS and datascience, preferably in an industry-backed research centre (think: Microsoft labs, or...
I've been heavily influenced by reading about the zettelkasten method (and its siblings); it seems to, theoretically, solve many problems that I (a bioinformatics student) encounter on a daily basis. However, I'm struggling with implementing said method in practice. While in biology I think the it will be easily applicable,...