Bayesian learning of noisy Markov decision processes

This work addresses the problem of estimating the optimal value function in a Markov Decision Process from observed state-action pairs. We adopt a Bayesian approach to inference, which allows both the model to be estimated and predictions about actions to be made in a unified framework, providing a principled approach to mimicry of a controller on the basis of observed data. A new Markov chain Monte Carlo (MCMC) sampler is devised for simulation from theposterior distribution over the optimal value function. This step includes a parameter expansion step, which is shown to be essential for good convergence properties of the MCMC sampler. As an illustration, the method is applied to learning a human controller.

Learning the dynamics of the external world: Brain inspired learning for robotic applications

Humans have exceptional abilities to learn new skills, manipulate tools and objects, and interact with our environment. In order to be successful at these tasks, our brain has become exceptionally well adapted to learning to deal not only with the complex dynamics of our own limbs but also with novel dynamics in the external world. While learning of these dynamics includes learning the complex time-varying forces at the end of limbs through the updating of internal models, it must also include learning the appropriate mechanical impedance in order to stabilize both the limb and any objects contacted in the environment. This article reviews the field of human learning by examining recent experimental evidence about adaptation to novel unstable dynamics and explores how this knowledge about the brain and neuro-muscular system can expand the learning capabilities of robotics and prosthetics. © 2006.

Bayesian Learning of Noisy Markov Decision Processes

We consider the inverse reinforcement learning problem, that is, the problem of learning from, and then predicting or mimicking a controller based on state/action data. We propose a statistical model for such data, derived from the structure of a Markov decision process. Adopting a Bayesian approach to inference, we show how latent variables of the model can be estimated, and how predictions about actions can be made, in a unified framework. A new Markov chain Monte Carlo (MCMC) sampler is devised for simulation from the posterior distribution. This step includes a parameter expansion step, which is shown to be essential for good convergence properties of the MCMC sampler. As an illustration, the method is applied to learning a human controller.

Risk-based approaches to sustainability in civil engineering

Infrastructure project sustainability assessment typically entails the use of specialised assessment tools to measure and rate project performance against a set of criteria. This paper looks beyond the prevailing approaches to sustainability assessments and explores sustainability principles in terms of project risks and opportunities. Taking a risk management approach to applying sustainability concepts to projects has the potential to reconceptualise decision structures for sustainability from bespoke assessments to becoming a standard part of the project decisionmaking process. By integrating issues of sustainability into project risk management for project planning, design and construction, sustainability is considered within a more traditional business and engineering language. Currently, there is no widely practised approach for objectively considering the environmental and social context of projects alongside the more traditional project risk assessments of time, cost and quality. A risk-based approach would not solve all the issues associated with existing sustainability assessments but it would place sustainability concerns alongside other key risks and opportunities, integrating sustainability with other project decisions.