I'm interested in how memory could be implemented in the brain and how memory content could be used to generalise across different and novel tasks. More specifically, some of the topics I'm currently exploring are Hopfield networks, episodic memory (from reinforcement learning literature), transfer learning (meta-learning, continual learning), hippocampus and entorhinal cortex models (grid-place cells, replay). The reason to review machine learning methods, particularly deep network architectures, as well as neuroscience, is because it seems like they share common share properties of biological and artificial systems that could be exploited to advance in both areas.