Download my PhD

Dr Edward Robinson

Computer Science Research Fellow

Hi, welcome to my academic site. In November 2011 I left the world of academia to work with a group of talented devlopers, marketers, and academics on applying interesting data-mining and machine learning techniques to the world of online retail. You can find more of me here.

Until October 2011, I was a post-doctoral Research Fellow at CERCIA, which is the University of Birmingham's centre of excellence for research in computational intelligence and applications, based within the School of Computer Science. My research and expertise span several areas: economic-inspired agent-based modelling; machine learning; and reputation and recommendation systems.

My PhD thesis, entitled: “Resource Allocation via Competing Marketplaces” was examined by Professor Dave Cliff and Dr John Bullinaria. The doctoral research specifically focussed on designing and analysing automated trader and market mechanisms for allocating multi-attribute computational resources. You can find out more by taking a look at the research section of this site.


  1. pdfE. Robinson, P. McBurney and X. Yao (2011). “Co-learning Segmentation in Marketplaces”. In P. Vrancx, M. Knudson and M. Grzes (Editors): Adaptive Learning Agents: ALA 2011 Workshop, 10th Autonomous Agents and Multi-agent Systems (AAMAS 2011), Taipei, Taiwan. Lecture Notes in Computer Science, Springer.
  2. pdfP. R. Lewis, A. Chandra, S. Parsons, E. Robinson, K. Glette, R. Bahsoon, J. Torresen and X. Yao (2011). “A survey of self-awareness and its application in computing systems”. In Proceedings of the first AWARE Workshop: Challenges in Achieving Self-Awareness in Autonomous Systems, Fifth IEEE International Conference on Self-Adaptive and Self-Organizing Systems (SASO 11), Ann Arbor, MI, USA.
  3. pdfE. Robinson, P. McBurney and X. Yao (2011). “Market Niching in Multi-attribute Computational Resource Allocation Systems”. In Proceedings of the 13th International Conference on Electronic Commerce (ICEC 11), Liverpool, UK.
  4. pdfE. Robinson, P. McBurney and X. Yao (2010). “How specialised are specialists? Generalisation properties of entries from the 2008 and 2009 TAC Market Design Competitions”. In E. David et al., Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms, volume 59 of Lecture Notes in Business Information Processing, Springer.
  5. pdfA. Brintrup, C. Davis, T. Gong, A. Ligtvoet, E. Robinson and W. Willigen (2009). “Distributed Control of Emergence: Local and Global Anti-component Strategies in Particle Swarms and Ant Colonies” In Proceedings of the third IEEE International Conference on Self-Adaptive and Self-Organizing Systems (SASO 09), San Francisco, CA, USA.
  6. pdfE. Robinson, P. McBurney and X. Yao (2009). “How specialised are specialists? Generalisation properties of entries from the 2008 TAC Market Design Competition”. In E. Gerding, editor, Proceedings of the 2009 Workshop on Trading Agent Design and Analysis (TADA ‘09), International Joint Conference in Artificial Intelligence (IJCAI 09), Pasadena, CA, USA.
  7. pdfE. Robinson and J. A. Bullinaria (2008). “Neuroevolution of Auto-Teaching Architectures.” In the 11th Neural Computation and Psychology Workshop (NCPW11), Oxford, UK, 16–18 July 2008.
  8. pdfE. Robinson, A. Channon and T. Ellis (2007). “Neuroevolution of Agents Capable of Reactive and Deliberative Behaviours in Novel and Dynamic Environments.” In Advances in Artificial Life: Proceedings of the 9th European Conference on Artificial Life (ECAL 2007), Lisbon, Portugal.


  1. pdfE. Robinson (2011). Resource Allocation via Competing Marketplaces. PhD dissertation, University of Birmingham, Birmingham, UK.

General Research Area

I'm interested in the design of economic mechanisms, whether automated trading strategies or market-mechanisms, particularly pertaining to the allocation or trading of resources via double auctions. Usually, this involves the application of novel machine learning approaches, for example unsupervised learning techniques such as reinforcement learning or evolutionary computation, to a variety of resource allocation problems.


Within the discipline of economics, the field of mechanism design concerns itself with the theoretical design of market-based mechanisms for resource allocation problems. Mechanism design treats the designing of mechanisms as setting the rules of a game for self-interested agents, such that when followed in a rational way, participants' interactions will lead to designer-desirable outcomes. And therein lies the rub. In most real-world settings, the design of appropriate mechanisms is non-trivial and complex—theoretical approaches are often not practical.

Rather, I treat the design of market and trader mechanisms as an engineering problem, where one wishes to strive towards some goal, e.g., economic efficiency or profit, while complying with practical constraints, e.g., minimising economic risk, communication overhead, or computational complexity.


I typically use the multi-agent systems paradigm to design and analyse these economic mechanisms. The MAS approach treats economic entities, e.g., traders and market institutions, as distinct software agents capable of independent reasoning and action. By using agent-based modelling approaches, very detailed and rich market simulations can be carried out, allowing one to analyse the behaviour and performance of different mechanisms in a systematic and repeatable way.

Tools of the trade

Designing automated mechanisms to either trade successfully or allocate resources efficiently is challenging in many real-world environments. Along the way, I have applied or have become familiar with, the following techniques/approaches:

  • Meta-heuristics such as Evolutionary Strategies and Genetic Algorithms;
  • Reinforcement learning algorithms such as N-armed Bandits, Q-learning and Temporal Difference Methods;
  • Statistical inference approaches such as Bayesian Statistics; and
  • Distributed trust and reputation systems.

Doctoral Research

My PhD dissertation may be downloaded here.


This thesis proposes a novel method for allocating multi-attribute computational resources via competing marketplaces. Trading agents, working on behalf of resource consumers and providers, choose to trade in resource markets where the resources being traded best align with their preferences and constraints. Market-exchange agents, in competition with each other, attempt to provide resource markets that attract traders, with the goal of maximising their profit.

Because exchanges can only partially observe global supply and demand schedules, novel strategies are required to automate their search for market niches. By applying a novel methodology, which is also used to explore, for the first time, the generalisation ability of market mechanisms, novel attribute-level selection (ALS) strategies are analysed in competitive market environments. Results from simulation studies suggest that using these ALS strategies, market-exchanges can seek out market niches under a variety of environmental conditions.

In order to facilitate traders’ selection between dynamic competing marketplaces, this thesis explores the application of a reputation system, and simulation results suggest reputation-based market-selection signals can lead to more efficient global resource allocations in dynamic environments. Further, a subjective reputation system, grounded in Bayesian statistics, allows traders to identify and ignore the opinions of those attempting to falsely damage or bolster marketplace reputation.


Until October 2011 I was a postdoctoral Research Fellow at the University of Birmingham, supported by a prestigious EPSRC PhD Plus fellowship, which I successfully applied for in 2010. Prior to that I successfully defended my PhD thesis, entitled “Resource Allocation via Competing Marketplaces”. And before all that I completed a Bachelor's degree in Artificial Intelligence and Computer Science in 2006, followed by a Master's degree in 2007 in Natural Computation, which I completed with distinction.

Over the last seven years of academic education and research, I've gained a strong understanding of several areas of computer science, but with particular expertise in areas of machine learning and artificial intelligence, such as Evolutionary Computation, Neural Networks, Reinforcement Learning and Agent-based Simulation. I also have a strong understanding of certain areas of Economics, such as Auction and Mechanism Design and Game Theory.

On the few occasions that my brain is not working on research problems, you will most likely find me either on the football pitch or in the pub.

Contact me

The best way to get in touch is by sending an email to ten.nosnibordde@em.

Social Media

You'll find me on LinkedIn,, and Mendeley.


School of Computer Science,
University of Birmingham,
B15 2TT,