Congratulations to two members of Computer Science academic staff, who have achieved success in the EPSRC First Grant scheme. The scheme, run by the Engineering and Physical Sciences Research Council (EPSRC), is there to help new academics apply for research funding at the start of their careers.
Both Dr Rob Alexander and Dr Stefano Pirandola have been awarded funding for their projects:
Dr Rob Alexander, Testing Autonomous Vehicle Software using Situation Generation
Autonomous vehicles (AVs) must be controlled by software, and such software thus has responsibility for safe vehicle behaviour. It is therefore essential that we rigorously test such software. This is difficult to do for AVs, as they have to respond appropriately to a great diversity of external situations as they go about their missions.
It is possible to find faults in an AV software specification by testing its behaviour in a variety of external situations, either in reality or in computer simulation. Such testing may reveal that the specification ignores certain situations (e.g. negotiating a motorway contraflow lane) or defines behaviour that is unsafe in a subset of situations (e.g. its policy for adapting to icy surfaces leads to unsafe speed control in crowded urban environments).
This project will test the hypothesis that testing based on coverage of possible external situations ("situation coverage") is an effective means of finding AV specification faults. We will test the hypothesis by creating a tool that generates situations for simulated AVs, both randomly and using heuristic search, and assessing whether higher situation coverage correlates with greater success at revealing seeded specification faults. (For the search, the fitness function will be based on the situation coverage achieved).
The project will draw on previous work on test coverage measures, on search-based testing, and on automated scenario generation in training simulations. To assess the effectiveness of the approach, we will use a small but practically-motivated case study of an autonomous ground vehicle, informed by the advice of an advisory panel set up for this project.
Dr Stefano Pirandola, Quantum Discrimination for Data Retrieval (qDATA)
Information is very important in our society. It is the "thing" which is processed by our computers and transmitted over the Internet. Every day we enjoy its benefits, since acquiring information means increasing our knowledge. For this reason, storing information is also very important. This is a process which is very common in our routine lives; for instance, think of a hard disk working in the background, or a DVD burned as a back-up of your most important data.
Advances in data storage could be much greater if they came from a deeper understanding of the concept of information. The basic unit of information, the bit, relies on our ability to distinguish between two states of a physical system. At the quantum level, storing and retrieving a bit of information relies on the capacity to discriminate between two quantum states of the system, e.g., spin up or spin down of an electron.
In our proposal, we consider a more advanced approach where information is encoded using quantum channels, i.e., the most general physical maps between quantum states. In our model, an encoder randomly picks a quantum channel from a pre-established ensemble, labeled by a classical variable. This channel is then stored in a black box and passed to a decoder. To identify the channel and retrieve the value of the variable, the decoder uses a transmitter, for feeding an input state into the box, and a receiver, for measuring the possible output states. Thus, data is stored in an ensemble of quantum channels and retrieved by the process of quantum channel discrimination.
Motivated by this approach, our first aim is to solve the general problem of quantum channel discrimination, by considering ensembles of Gaussian channels and assuming decoders with limited energy. This is an open problem, whose optimal solution will provide the core for a general theory on Gaussian channel discrimination.
This theory will then be applied to practical scenarios which are important for data storage. We will consider the quantum reading of digital memories, where the use of faint quantum light is remarkably efficient in retrieving data from classical optical discs (resembling CDs and DVDs). Our aim is to optimize this model by including error correcting codes and, most importantly, to make it practical by studying all the details of its optical implementation, where the inevitable presence of diffraction causes effects of inter-bit interference. Thanks to this study, we will be able to promote this theoretical idea to the level of a technological prototype, ready to be experimentally implemented.
The field implementation of a quantum reader could be a breakthrough in data storage, since we could increase data transfer rates and storage capacities of our digital memories by orders of magnitude. Furthermore, thanks to the non-invasive nature of the quantum light, new photo-degradable materials could be used by the industry for the construction of new types of organic memories. Our approach is high-risk but it could open the way to radically new forms of information technologies.
Then, a generalization of quantum reading is quantum pattern recognition. Here we aim to prove how quantum correlations can dramatically improve the performances of pattern matching in supervised and unsupervised algorithms (for instance, for data clustering). Quantum pattern recognition can potentially lead to a dramatic boost in the classification of raw data with minimal use of energy, negligible error rates and fast acquisition times. This technique could be used for the probing of very fragile biological or human samples in order to recognize the presence of bacterial growths or cancerous cells. Thanks to its non-invasive nature, quantum light could also be used for a continuous real-time probing of such samples. Results could be revolutionary in the long-term, providing completely new techniques for biological analysis and medical imaging.