No description available.

No description available.

Whether it is the accumulation of capital or capabilities that accounts for rapid development of Newly Industrialized Countries (NICs) has been a focus of debate. The former, which informs approaches to development often adopted by international agencies, reflects neoclassical perspectives. The later, by contrast, reflects evolutionary approaches with deliberate learning and innovation as dominant factors. The purpose of this thesis is to understand development by exploring how it is influenced by learning and innovation focusing on the factors, mechanisms and institutional conditions that foster learning and innovation in Malawi. This thesis has adopted quantitative and qualitative methods informed extensively by theoretical perspectives on the knowledge, learning, innovation and development nexus. Using primary survey and secondary data, a conceptual framework that emerged from contrasting perspectives on theories of the firm has situated a quantitative understanding of how firms in Malawi learn and innovate and the impact of institutional conditions. A qualitative approach, however, has enabled identification of underlying mechanisms that foster learning and innovation thus, providing bases for articulating how evolutionary perspectives can enhance Malawi's development prospects. The thesis finds that successful development is conditioned on understanding it as an interactive process of learning and innovation hinged on addressing systemic failure regarding acquisition and utilization of knowledge by producers, firms especially. We find that failures related to institutional conditions on market and social capabilities, governance and communication and knowledge infrastructure have created a business environment in Malawi that does not foster firm learning and innovation. Thus, firms are more inclined to exploiting existing capabilities leading to static rather than dynamic efficiency. This behaviour reflects dominance of neoclassical perspective of development by stakeholders. We find, therefore, that Malawi's development prospects are predicated on innovation in the delivery of knowledge-related services to producers thus, innovation in public goods. At firm-level, action that: promotes firms' investment in on-the-job training, engenders dialogue, fosters collaboration; and builds knowledge stock positively influences learning and innovation capability. We find that high learning firms, under entrepreneurial leadership, exemplify an evolutionary understanding of the role of knowledge in production. They deliberately foster these behavioural and cognitive factors for which they are rewarded with superior performance. At national level, we find that contrary to neoclassical perspectives, Malawi's development is conditioned on purposive action by all stakeholders, government in particular, to mitigate constraints on learning and innovation arising from idiosyncratic aspects of the business environment. This evolutionary perspective entails entrepreneurial leadership in government and adoption of a national learning and innovation system approach to development. We argue that building coalitions focused on fostering knowledge flows to firms, especially those in the manufacturing sector which we find to be the basis for structural change of the economy, is a necessary though not sufficient pre-condition for Malawi's development.

This thesis was inspired by the difficulties of artificial evolutionary systems in finding elegant and well structured, regular solutions. That is that the solutions found are usually highly disorganized, poorly structured and exhibit limited re-use, resulting in bloat and other problems. This is also true of previous developmental evolutionary systems, where structural regularity emerges only by chance. We hypothesise that these problems might be ameliorated by incorporating repeated evaluations on increasingly difficult problems in the course of a developmental process. This thesis introduces a new technique for learning complex problems from a family of structured increasingly difficult problems, Evolutionary Developmental Evaluation (EDE). This approach appears to give more structured, scalable and regular solutions to such families of problems than previous methods. In addition, the thesis proposes some bio-inspired components that are required by developmental evolutionary systems to take full advantage of this approach. The key part of this is the developmental process, in combination with a varying fitness function evaluated at multiple stages of development, generates selective pressure toward generalisation. This also means that parsimony in structure is selected for without any direct parsimony pressure. As a result, the system encourages the emergence of modularity and structural regularity in solutions. In this thesis, a new genetic developmental system called Developmental Tree Adjoining Grammar Guided Genetic Programming (DTAG3P), is implemented, embodying the requirements above. It is tested on a range of benchmark problems. The results indicate that the method generates more regularly-structured solutions than the competing methods. As a result, the system is able to scale, at least on the problem classes tested, to very complex instances the system encourages the emergence of modularity and structural regularity in solutions. In this thesis, a new genetic developmental system called Developmental Tree Adjoining Grammar Guided Genetic Programming (DTAG3P), is implemented, embodying the requirements above. It is tested on a range of benchmark problems. The results indicate that the method generates more regularly-structured solutions than competing methods. As a result, the system is able to scale, at least on the problem classes tested, to very complex problem instances.

The aim of this work is to design and implement a method for the evolutionary design of neural networks with generative encoding. The proposed method is based on J. F. Miller's approach and uses a brain model that is gradually developed and which allows extraction of traditional neural networks. The development of the brain is controlled by programs created using cartesian genetic programming. The project was implemented in Python with the use of Numpy library. Experiments have shown that the proposed method is able to construct neural networks that achieve over 90 % accuracy on smaller datasets. The method is also able to develop neural networks capable of solving multiple problems at once while slightly reducing accuracy.