I am a Fellow and Tutor in Physics and also Director of Studies for this subject. I am teaching maths to first year physics undergraduates and lecture on Quantum Communication and Information in the Physics department. My research group is located at the Clarendon Laboratory and our main research interests lie in the quantum physics of ultracold atomic gases.
My research is currently focused on the theoretical study of strongly correlated quantum systems which are periodically driven and far away from equilibrium. These unusual physical states can for instance be created in the laboratory by shining laser light onto an ultra-cold ensemble of neutral atoms. The coherence times of these atoms are sufficiently long for purely quantum many-body phenomena to become visible in the experiments. New insights of fundamental importance into the nature of quantum physics and in particular the properties and structure of entanglement can be investigated in these experiments. Another option to study quantum states far from equilibrium is to excite condensed matter (e.g. specially designed organic salts) using THz radiation on an ultra-fast time scale. In this situation quantum evolution becomes visible at room temperature, albeit at much shorter time scales. Effects observed in such experiments may enhance future technology by exploiting quantum dynamical processes.
You can find out more about my work by reading the following articles here on the ASC website:
- Tensor Networks in a Nutshell: http://www.keble.community.librios.com/?id=988
- Tensor Network Algorithms: A new paradigm for modelling complex systems and analysing data: http://www.keble.community.librios.com/?id=-423&cid=1002
RESEARCH FOCUS: TENSOR NETWORK ALGORITHMS (TNAs)
A major part of our research efforts as a part of the Keble Networks cluster is to apply the methods and tools from physics described in the articles above above to wider settings. We are currently exploring this in the context of Markov chains for classical stochastic systems such as queues, production lines and simple models of vehicular traffic. To gain traction in this area, TNA need to compete with Monte-Carlo methods. Preliminary results suggest that TNAs can describe the probability of extremely rare events far better than Monte-Carlo sampling. This is thought to be because of the unique way in which tensor networks approximate the full multi-variant probability distribution describing Markov chain systems. If proven, this could lead to a number of killer applications of TNA in this area.