Biography

I am a research scientist at DeepMind. My current research focus is on building robust and verifiable AI systems that can be trusted to behave reasonably even under adversarial circumstances (see this talk and these slides for a summary of my recent research).

More generally, I am interested in developing algorithms to guarantee correct behavior in complex autonomous systems. My PhD thesis focused on efficient algorithms for stochastic optimal control and reinforcement learning in risk-sensitive situations (see this and this). I also worked on applying ideas from robust and risk averse control to electric power grids with high penetrations of uncertain renewable generation sources like wind and solar energy (see this and this ), before moving to DeepMind.

My research has received several best paper awards at AI conferences ( UAI 2018, CP 2016 , UAI 2014 and ECML 2008). My work at DeepMind has also been incorporated into AI-based battery management solutions for android phones.

• dvij@cs.washington.edu
• 6 Pancras Square, N1C 4AG, London, UK