Hi, I’m Akbir. I am an artificial intelligence (AI) PhD candidate at UCL, advised by Tim Rocktäschel and Edward Grefenstette and affiliated with UCL DARK lab.
I’m interested in studying techniques to ensure cooperative AI. I consider these problems primarily in the setting of multi-agent deep reinforcement learning.
In my extra time, I help as a Senior Researcher at Tractable, where I develop multi-modal models to evaluate damage on vehicles. Before this, I was CRO of Spherical Defence Labs, using language models to create network-level intrusion detection systems.
Previously, I received an MPhil in Computer Science from the University of Cambridge, where I had the pleasure of working with Hatice Gunes. I completed my undergraduate in Mathematics and Physics at UCL, spending a year abroad at the University Of Toronto. For a more in-depth account, see my resume.
If you care about the same things, feel free to drop me an email.