Exploration vs exploitation in differential evolution

Differential Evolution (DE) is a tool for efficient optimisation, and it belongs to the class of evolutionary algorithms, which include Evolution Strategies and Genetic Algorithms. DE algorithms work well when the population covers the entire search space, and they have shown to be effective on a large range of classical optimisation problems. However, an undesirable behaviour was detected when all the members of the population are in a basin of attraction of a local optimum (local minimum or local maximum), because in this situation the population cannot escape from it. This paper proposes a modification of the standard mechanisms in DE algorithm in order to change the exploration vs. exploitation balance to improve its behaviour.

The transformation between exploration and exploitation applied to inventors of packaging innovations

While the inventor is often the driver of an invention in the early stages, he/she needs to move between different social networks for knowledge in order to create and capture value. The main objective of this research is to propose a literature-based framework based on innovation network theory and complemented with C-K theory, in order to analyze the invention/innovation process of inventors and the product concepts in a packaging industry context. Empirical input from three case studies of packaging inventions and their inventors is used to elaborate the suggested framework.The article identifies important gaps in the literature of innovation networks. This is addressed through a theoretical framework based on network theories, complemented with C-K theory for the product design level. The strength-of-ties dimension of the theoretical framework suggests, in agreement with the mainstream literature and the cases presented, that weak ties are required to access the knowledge related to exploration networks and strong ties are required to utilize the knowledge in the exploitation network. The transformation network is an intermediate step acting as a bridge where entrepreneurs can find required knowledge. The transformation network is also an intermediate step where entrepreneurs find financing and companies interested in commercializing inventions. (C) 2010 Elsevier Ltd. All rights reserved.

Balancing Exploration and Exploitation: A New Algorithm for Active Machine Learning

Active machine learning algorithms are used when large numbers of unlabeled examples are available and getting labels for them is costly (e.g. requiring consulting a human expert). Many conventional active learning algorithms focus on refining the decision boundary, at the expense of exploring new regions that the current hypothesis misclassifies. We propose a new active learning algorithm that balances such exploration with refining of the decision boundary by dynamically adjusting the probability to explore at each step. Our experimental results demonstrate improved performance on data sets that require extensive exploration while remaining competitive on data sets that do not. Our algorithm also shows significant tolerance of noise.

U.S. Antarctic policy : Hearing...94th Congress, 1st Session on U.S. policy with respect to mineral exploration and exploitation in the Antarctic /

Executive hearing held May 15, 1975; made public July 6, 1975

Dynamic client portfolios as sources of ambidexterity:exploration and exploitation within and across client relationships

This paper explains how dynamic client portfolios can be a source of ambidexterity (i.e., exploration and exploitation) for knowledge-intensive firms (KIFs). Drawing from a unique qualitative dataset of firms in the global reinsurance market, we show how different types of client relationships underpin a dynamic client portfolio and become a source of ambidexterity for a KIF. We develop a process model to show how KIFs attain knowledge by segmenting their client portfolios, use that knowledge to explore and exploit within and across their client relationships, and dynamically adjust their client portfolios over time. Our study contributes to the literature on external sources of ambidexterity and dynamic management of client knowledge within KIFs.

The lost experiment in exploration and exploitation

This paper focuses on James March’s 1991 article on ‘Exploration and Exploitation in Organizational Learning’, which is now the seventh most highly cited paper in management and organisation studies. March’s paper is based on a computer program that simulates the collective and individual learning of a group of fifty individuals. The largely forgotten story that this paper re-calls is the real-life experiment that March, in large part, designed and conducted when he was the new ‘boy Dean’ of the School of Social Sciences in the University of California at Irvine between 1964 and 1969. Taken together, both stories illuminate important moments in the history of organisation studies. The comparison suggests that March’s model, which was probably the first simulation of an organisation learning, also worked to constitute rather than model the phenomenon.

Impacts of oil exploration and exploitation activities on marine environment

Iran possess huge oil reservoir and occupies second place in OPEC. Recent investigation has revealed that reservoir capacity in the country amount to 60 billion barrel of oil. Many measures has been carried out to increase production capacity of oil fields to 4.2 million barrel per day. Thus any distribution in oil exploration may leave adverse effects on social and economic activities. Unfortunately due to absence of a comprehensive CPM on environmental impact assessment, lots of environmental distribution has been occurred in land and off-shore. It is well known that implementation of EIA can reduce environmental hazards. In the present investigation, all major and minor activities associated with oil exploration is identified and subsequently their effects on physical, chemical and biological environment (aquatic) has been brought out. In this context, economical, social and cultural effects of marine oil exploration is also discussed. Subsequently, all methods of EIA were studied and best mitigation plans were drawn up both for exploitation and exploration phases.

War and Natural Resource Exploitation

We build a theoretical framework that allows for endogenous conflict behaviour (i.e., fighting efforts) and for endogenous natural resource exploitation (i.e., speed, ownership, and investments). While depletion is spread in a balanced Hotelling fashion during peace, the presence of conflict creates incentives for rapacious extraction, as this lowers the stakes of future contest. This voracious extraction depresses total oil revenue, especially if world oil demand is relatively elastic and the government's weapon advantage is weak. Some of these political distortions can be overcome by bribing rebels or by government investment in weapons. The shadow of conflict can also make less efficient nationalized oil extraction more attractive than private extraction, as insecure property rights create a holdup problem for the private firm and lead to a lower license fee. Furthermore, the government fights less intensely than the rebels under private exploitation, which leads to more government turnover. Without credible commitment to future fighting efforts, private oil depletion is only lucrative if the government's non-oil office rents are large and weaponry powerful, which guarantees the government a stronger grip on office and makes the holdup problem less severe.