Hello, I'm a graduate student in the CoMPLEX program at UCL, where I am currently working on my PhD jointly at the Gatsby Computational Neuroscience Unit and Sainsbury Wellcome Centre for Neural Circuits and Behavior. I work with Peter Latham and Adam Kampff.
I'm interested in understanding neural computation. I ask questions like:
- Why are brains so good at certain types of computation? What is it about problems that the brain solves effortlessly (e.g. vision) that makes them so difficult for digital computers?
- What kinds of neural circuit architectures enable useful computations?
- Which biological properties endow the brain its computational power?
- Can we exploit these to improve our own machine learning algorithms?
I take theoretical approaches to answering such questions, with an emphasis on producing results that can be tested in the lab. After all, I want to understand the brain, not a neural net. My PhD thesis largely involves developing biologically plausible neural networks that do useful things. As neurotechnological advances allow us to examine neural circuits and dynamics at finer scales, complementary progress in the theory of neural computation will allow us to interpret their implications for the brain's computational capabilities. On this front, we can borrow ideas from modern machine learning algorithms, many of which were originally inspired by neuronal architectures but have long since abandoned their biological roots. My work aims to bring together biology and computation, with the hope of yielding insights in both fields.