Despite the increasing relevance of knowledge-based activities in the global economy, their spatial distribution at the intrametropolitan scale has rarely been studied. The aim of this paper is to shed light on the intrametropolitan spatial localisation of knowledge-based activities by assessing some of its main causal mechanisms, paying special attention to agglomeration economies and borrowed size (both questioned in the era of Information and Communications Technologies because of the theorised 'death of distance'), the controversial role of industrial diversity and specialisation (traditionally considered exclusive processes) and path dependence.The empirical application refers to the case of the Madrid urban region for the period between 2012 and 2017. This study area proves particularly appropriate for two main reasons. First, the area is a main urban region in the European and global urban systems. Second, the area is a good example of a multicore urban region with a traditionally strong metropolis. The results show that the knowledge economy remains strongly dependent on the size of the city itself (agglomeration economies) and of the surrounding areas (borrowed size), that the specialisation and diversity processes are complementary and appear at different scales and that, in the specialisation of certain areas in some knowledge economy activities, the city's trajectory (path dependence) plays an important role. The results of the present study have various practical implications for policymakers when considering not only strategies (and investments) to coordinate land availability, worker training, connections between research centres and companies and amenities but also strategies to coordinate with the surrounding municipalities (e.g. in terms of mobility or joint public and private projects).

In this paper we deal with the problem of automatically optimizing the gait of a robot for forward walking speed. Each different walking surface and/or the wear and tear of the robots determines the speed of the robot. This means that a specific gait for one surface may not be valid on another surface or even on the same surface some time later. Given a parametrized walk designed for the robot and one specific robot, we approach the problem of searching over the space of gaits for this robot by using an estimation of distribution algorithm and our knowledge of the problem, which allows us to considerably reduce the space of gaits over which the algorithm searches. We have implemented and tested our method using the Sony AIBO ERS-7 robot, significantly improving on our previous gait in a short period of training time.