Intelligence Augmentation Research Articles

Big Data breaking barriers - first steps on a long trail

Abstract. Most data sets and streams have a geospatial component. Some people even claim that about 80% of all data is related to location. In the era of Big Data this number might even be underestimated, as data sets interrelate and initially non-spatial data becomes indirectly geo-referenced. The optimal treatment of Big Data thus requires advanced methods and technologies for handling the geospatial aspects in data storage, processing, pattern recognition, prediction, visualisation and exploration. On the one hand, our work exploits earth and environmental sciences for existing interoperability standards, and the foundational data structures, algorithms and software that are required to meet these geospatial information handling tasks. On the other hand, we are concerned with the arising needs to combine human analysis capacities (intelligence augmentation) with machine power (artificial intelligence). This paper provides an overview of the emerging landscape and outlines our (Digital Earth) vision for addressing the upcoming issues. We particularly request the projection and re-use of the existing environmental, earth observation and remote sensing expertise in other sectors, i.e. to break the barriers of all of these silos by investigating integrated applications.

An intelligent augmentation of particle swarm optimization with multiple adaptive methods

Over the last two decades, the newly developed optimization technique – Particle Swarm Optimization (PSO) has attracted great attention. Two common criticisms exist. First, most existing PSOs are designed for a specific search space thus an algorithm performing well on a diverse set of problems is lacking. Secondly, PSO suffers premature convergence. To address the first issue, we propose to augment PSO via the fusion of multiple search methods. An intelligent selection mechanism is developed based on an effectiveness index to trigger appropriate search methods. In this research, two search techniques are studied: a non-uniform mutation-based method and an adaptive sub-gradient method. We further improve the proposed PSO using adaptive Cauchy mutation to prevent premature convergence. As a result, an augmented PSO with multiple adaptive methods (PSO-MAM) is proposed. The performance of PSO-MAM is tested on 43 functions (uni-modal, multi-modal, non-separable, shifted, rotated, noisy and mis-scaled functions). The results are compared in terms of solution quality and convergence speed with 10 published PSO methods. The experimental results demonstrate PSO-MAM outperforms the comparison algorithms on 36 out of 43 functions. We conclude, while promising, there is still room for improving PSO-MAM on complex multi-modal functions (e.g., rotated multi-modal functions).

Robotic and artificial intelligence for keyhole neurosurgery: The ROBOCAST project, a multi-modal autonomous path planner

The robot and sensors integration for computer-assisted surgery and therapy (ROBOCAST) project (FP7-ICT-2007-215190) is co-funded by the European Union within the Seventh Framework Programme in the field of information and communication technologies. The ROBOCAST project focuses on robot- and artificial-intelligence-assisted keyhole neurosurgery (tumour biopsy and local drug delivery along straight or turning paths). The goal of this project is to assist surgeons with a robotic system controlled by an intelligent high-level controller (HLC) able to gather and integrate information from the surgeon, from diagnostic images, and from an array of on-field sensors. The HLC integrates pre-operative and intra-operative diagnostics data and measurements, intelligence augmentation, multiple-robot dexterity, and multiple sensory inputs in a closed-loop cooperating scheme including a smart interface for improved haptic immersion and integration. This paper, after the overall architecture description, focuses on the intelligent trajectory planner based on risk estimation and human criticism. The current status of development is reported, and first tests on the planner are shown by using a real image stack and risk descriptor phantom. The advantages of using a fuzzy risk description are given by the possibility of upgrading the knowledge on-field without the intervention of a knowledge engineer.

In Love with our Technology: Virtual Reality

The intellectual history suggests that desire for intelligence augmentation underlies the idea of virtual reality (VR). The assumption underlying this teleological thrust towards intelligence augmentation is that intellectual advances are made by improving and facilitating human thinking and problem solving. This was thought possible only if computers were integrated effectively into the whole ecology of thought by virtue of a human-computer partnership, or symbiosis. VR media are seen as augmenting intelligence by more direct, intuitive channelling of information through the senses to establish a more direct interface, or experience with knowledge. The history of the idea of virtuality shows that policy, personnel choice and placement, and research and development aimed to implement this vision for VR. Related work led to the internet as focus was placed on networked, cooperative interaction - computers and people working together to augment human intelligence.

A user‐centred approach to the design of an expert system for training

This paper reports the objectives, design approach and first results of a research project that aims to develop an intelligent computer‐based learning environment for industrial applications. Jonas, an expert system, is part of a modelling/simulation environment which enables shopfloor workers to test and to put new philosophies of work into practice in the context of manufacture. Our approach is focused on three main ideas: the Intelligence Augmentation paradigm, user‐centred system design and the constructionist theory of learning. The architecture of the modelling environment and the main features of Jonas are presented as an illustration of the design ideas discussed in the paper.

Thinking with machines: Intelligence augmentation, evolutionary epistemology, and semiotic

Thinking with machines: Intelligence augmentation, evolutionary epistemology, and semiotic