Corrective Advice Research Articles

Expert advisory system determining the functional reliability of the human operator based on psychological and psychophysiological indicators

The article is dedicated to the formation of a "generalized portrait" of the condition of the human operator, which is the main component of the ergatic system, based on psychophysiological indicators, and on this basis, providing corrective advice through an expert-advisory system. A formal-constructive model "generalized portrait" is created according to the stratified description of psychological and psychophysiological indicators. Each indicator included in the strats is realized by means of the corresponding software. Each component of the "generalized portrait" is assigned to different scale types (ratio, interval, ordinal, nominative). The theory of fuzzy sets and relations is used to unify different types of scales. The decision and advice functional block of the expert-advisory system makes a decision about the functional reliability based on the fuzzy knowledge based on the current state of the human-operator.

Reconceptualising Afrophobia in post-apartheid South Africa: a conflict transformational perspective

Xenophobic violence has assumed a regular appearance in the body politic of South Africa. Such xenophobia/racism is mainly targeted at migrants from elsewhere in Africa. The phenomenon flies in the face of expectations, given the history of the struggle against apartheid, and the supportive role of the frontline states in support of that struggle. It runs counter to the inclusive ethos of the ANC government, the Rainbow Nation and South Africa’s progressive constitution. This contribution, in the spirit of conflict transformation, focuses on the occurrence of xenophobia in South Africa. Given the complexity of the context, it seeks to explain the phenomenon, using a selection of commonly used theoretical approaches, mediated through the lenses of the Just World theory, and the notion of state crimes. It challenges any notion of South Africa as a post-conflict society, finds significant government culpability for the episodes, and attempts to provide corrective policy advice, using a dynamic systems theoretical approach.

Reinforcement learning of motor skills using Policy Search and human corrective advice

Robot learning problems are limited by physical constraints, which make learning successful policies for complex motor skills on real systems unfeasible. Some reinforcement learning methods, like Policy Search, offer stable convergence toward locally optimal solutions, whereas interactive machine learning or learning-from-demonstration methods allow fast transfer of human knowledge to the agents. However, most methods require expert demonstrations. In this work, we propose the use of human corrective advice in the actions domain for learning motor trajectories. Additionally, we combine this human feedback with reward functions in a Policy Search learning scheme. The use of both sources of information speeds up the learning process, since the intuitive knowledge of the human teacher can be easily transferred to the agent, while the Policy Search method with the cost/reward function take over for supervising the process and reducing the influence of occasional wrong human corrections. This interactive approach has been validated for learning movement primitives with simulated arms with several degrees of freedom in reaching via-point movements, and also using real robots in such tasks as “writing characters” and the ball-in-a-cup game. Compared with standard reinforcement learning without human advice, the results show that the proposed method not only converges to higher rewards when learning movement primitives, but also that the learning is sped up by a factor of 4–40 times, depending on the task.

Examining the validity and reliability of a portable sleep posture assessment protocol, using infrared cameras, under a variety of light and bed cover situations in the home environment.

Spinal symptoms of pain and stiffness on waking have been linked to sleep posture. Sleep posture is commonly classified as supine, side lying and prone. It is clinically postulated that sleeping postures with sustained end of range rotation and extension may influence pain sensitive spinal tissues. However, the lack of a valid and reliable method of assessing sleep posture, means clinicians are unable to provide corrective advice based upon evidenced based research. To determine the validity and reliability of a sleep posture recording protocol in the home environment. Twenty health professionals viewed a pre-recorded video recording of randomised sleep postures under natural and infrared light situations, with a variety of bed coverings, to represent the habitual environment. Sleep postures were classified into six categories including two intermediate postures (supported side lying and provocative side lying). Viewing was repeated after two days. Intra-and inter-rater reliability were excellent; Cohen's Kappa = .93 (95% CI 0.80 to 1.0) and Fleiss Kappa = 0.83 (95% CI 0.82 to 0.84) respectively. Validity, determined as concordance between the health professionals' classifications and the known postures, was also excellent Cohen's Kappa = .91 (95% CI 0.77 to 1.0). Reliable and valid assessment of sleep posture, including intermediate postures, could be achieved using low cost, portable, infrared video recording equipment, under a variety of lighting conditions and a variety of bed cover situations typical of the home environment.

An Interactive Framework for Learning Continuous Actions Policies Based on Corrective Feedback

The main goal of this article is to present COACH (COrrective Advice Communicated by Humans), a new learning framework that allows non-expert humans to advise an agent while it interacts with the environment in continuous action problems. The human feedback is given in the action domain as binary corrective signals (increase/decrease the current action magnitude), and COACH is able to adjust the amount of correction that a given action receives adaptively, taking state-dependent past feedback into consideration. COACH also manages the credit assignment problem that normally arises when actions in continuous time receive delayed corrections. The proposed framework is characterized and validated extensively using four well-known learning problems. The experimental analysis includes comparisons with other interactive learning frameworks, with classical reinforcement learning approaches, and with human teleoperators trying to solve the same learning problems by themselves. In all the reported experiments COACH outperforms the other methods in terms of learning speed and final performance. It is of interest to add that COACH has been applied successfully for addressing a complex real-world learning problem: the dribbling of the ball by humanoid soccer players.

El gobierno del cuerpo

Identified as a carnavalesque space for the subversion of social codes and for the projection of personal anxieties, Baltasar del Alcázar's (1530–1606) joyous poetry is built at the margins of authorized medical discourses. Colloquial in verse and tone, the poem “Cena jocosa” contests corrective and disciplinarian advice contained in its contemporary health guides, proposing a carefree enjoyment of the table. This happens at a time in the history of medicine when dietary choices were endowed with a profound ethical sense, and also at a time in the history of the Spanish empire when the bodily government announced models for state administration and patterns for rhetorical assimilation. This article explores the poem “Cena jocosa” in light of the hygienic discourse of health guides and, by extension, the late medieval regimina principorum and classical symposiac literature, the two poetical traditions that nourished Renaissance health guides.

Critiquing contractors' scheduling by integrating rule-based and case-based reasoning

Nowadays, projects involve many activities. Hence, the review process requires the scheduler to be experienced both in field operations and in computer scheduling software. However, the dilemma faced by contractors is that many senior project managers are experienced in the field but not in computers, whereas many young engineers are skilled in computers but not in field operations. This work proposes a generalized framework to represent schedule knowledge, and a computer-based system that analyzes a project schedule and offers corrective advice on potential errors by integrating case-based and rule-based reasoning.

Online preventive dynamic security of isolated power systems using decision trees

In this paper, the application of decision trees (DTs) for online preventive dynamic security of isolated power systems is presented. DTs provide the necessary computational speed for online performance and the flexibility of providing preventive control. Emphasis is placed on the online use of the DTs to test the dynamic security of each generation dispatch scenario and thus to provide corrective advice via generation redispatch. Moreover, the algorithm implemented provides the flexibility of displaying the cost of each redispatch. In this way, the method can help in objective decision-making. Results from the application of the system on actual load series from the island of Crete, where the proposed system is in trial operation, are presented.

A multi-level approach to the detection of second language learners errors

A natural language generation system used for computer-aided language learning generates stimuli, or 'questions', along with one or more anticipated learner responses. To provide the learner with intelligent feedback, the system initially compares the actual responses entered by the learner with those produced by the system. The differences discovered in this way may be though of as symptoms; the eventual aim is to diagnose the deeper problems, misunderstandings of the language that underlie the symptoms, to offer corrective advice. This paper considers alternative approaches to the comparison process, and describes a two-level algorithm, based on approximate string matching, which determines and categorizes the differences between learner and system responses. The process detects variations in word order, insertions and deletions. It also reports on the use of synonyms or malwords, errors in morphology, phonological errors and typing slips. In subsequent work, these will form the basis for the adaptive generation of new stimuli, as well as for the deeper diagnosis of learner difficulties. Some typical results are shown.

A knowledge-based advisory system for statistical quality control

Implementation of an effective quality management system requires the ready availability of expert statistical quality control (SQC) practitioners. However, this expertise may be unavailable to many small- and medium-size manufacturing organizations. Knowledge-based systems (KBS) can be used to make SQC expertise easily available to these organizations. This paper describes a KBS that has been developed for providing assistance in: (1) the selection and design of appropriate quality control charts; (2) the process monitoring analyses; and (3) providing corrective advice based on the monitoring analyses results. The KBS runs on a microcomputer and has been developed using a commercially available development shell.

Spacelab life sciences-1 electrical diagnostics expert system

The Spacelab Life Sciences-1 (SLS-1) Electrical Diagnostic (SLED) expert system is a continuous, real-time knowledge-based system to monitor and diagnose electrical system problems in the Spacelab. After fault isolation, the SLED system provides corrective procedures and advice to the ground-based console operator. The SLED system uses the Unit (frame) of KEE to represent the knowledge about the electrical components and uses KEE-Bit-maps to represent the electrical schematics. The diagnostic logic, stored as a set of LISP structures, mirrors that in the malfunction procedures defined in the Spacelab Flight Data File Malfunction Procedures Handbook (JSC-18927) of NASA. The SLED system utilizes downlink telemetry data as input. The system performs some initial screening of the data in order to recognize patterns representing serious problems, and updates its knowledge about the status of Spacelab every 3 seconds. Important parameters are monitored via Active-Values within KEE. The Active-Value kicks off the diagnostic analysis to determine the source of the problems if any problem has been identified. The system supports multiprocessing of malfunctions and allows multiple failures to be handled simultaneously. The user can examine each of the reported problems and receive corrective advice related to each problem. Information which is readily available via a mouse click includes: general information about the system and each component, the electrical schematics, the recovery procedures of each malfunction, and an explanation of the diagnosis. A rich set of user interfaces is provided in SLED. Various tools have been included which allow a non-programmer to define new diagnostic procedures, define and update schematics of the electrical system, and to change the SLED model by changing the graphical representation. Each tool and function has explanatory prompts to aid the user.