Mind change complexity of learning logic programs

The present paper motivates the study of mind change complexity for learning minimal models of length-bounded logic programs. It establishes ordinal mind change complexity bounds for learnability of these classes both from positive facts and from positive and negative facts. Building on Angluin’s notion of finite thickness and Wright’s work on finite elasticity, Shinohara defined the property of bounded finite thickness to give a sufficient condition for learnability of indexed families of computable languages from positive data. This paper shows that an effective version of Shinohara’s notion of bounded finite thickness gives sufficient conditions for learnability with ordinal mind change bound, both in the context of learnability from positive data and for learnability from complete (both positive and negative) data. Let Omega be a notation for the first limit ordinal. Then, it is shown that if a language defining framework yields a uniformly decidable family of languages and has effective bounded finite thickness, then for each natural number m >0, the class of languages defined by formal systems of length <= m: • is identifiable in the limit from positive data with a mind change bound of Omega (power)m; • is identifiable in the limit from both positive and negative data with an ordinal mind change bound of Omega × m. The above sufficient conditions are employed to give an ordinal mind change bound for learnability of minimal models of various classes of length-bounded Prolog programs, including Shapiro’s linear programs, Arimura and Shinohara’s depth-bounded linearly covering programs, and Krishna Rao’s depth-bounded linearly moded programs. It is also noted that the bound for learning from positive data is tight for the example classes considered.

Evidence of disagreement between patient-perceived change and conventional longitudinal evaluation of change in health-related quality of life among older adults

Objective: To identify agreement levels between conventional longitudinal evaluation of change (post–pre) and patient-perceived change (post–then test) in health-related quality of life. Design: A prospective cohort investigation with two assessment points (baseline and six-month follow-up) was implemented. Setting: Community rehabilitation setting. Subjects: Frail older adults accessing community-based rehabilitation services. Intervention: Nil as part of this investigation. Main measures: Conventional longitudinal change in health-related quality of life was considered the difference between standard EQ-5D assessments completed at baseline and follow-up. To evaluate patient-perceived change a ‘then test’ was also completed at the follow-up assessment. This required participants to report (from their current perspective) how they believe their health-related quality of life was at baseline (using the EQ-5D). Patient-perceived change was considered the difference between ‘then test’ and standard follow-up EQ-5D assessments. Results: The mean (SD) age of participants was 78.8 (7.3). Of the 70 participants 62 (89%) of data sets were complete and included in analysis. Agreement between conventional (post–pre) and patient-perceived (post–then test) change was low to moderate (EQ-5D utility intraclass correlation coefficient (ICC)¼0.41, EQ-5D visual analogue scale (VAS) ICC¼0.21). Neither approach inferred greater change than the other (utility P¼0.925, VAS P¼0.506). Mean (95% confidence interval (CI)) conventional change in EQ-5D utility and VAS were 0.140 (0.045,0.236) and 8.8 (3.3,14.3) respectively, while patient-perceived change was 0.147 (0.055,0.238) and 6.4 (1.7,11.1) respectively. Conclusions: Substantial disagreement exists between conventional longitudinal evaluation of change in health-related quality of life and patient-perceived change in health-related quality of life (as measured using a then test) within individuals.

Analysis of HapMap tag-SNPs in dysbindin (DTNBP1) reveals evidence of consistent association with schizophrenia

Dystrobrevin binding protein 1 (DTNBP1), or dysbindin, is thought to be critical in regulating the glutamatergic system. While the dopamine pathway is known to be important in the aetiology of schizophrenia, it seems likely that glutamatergic dysfunction can lead to the development of schizophrenia. DTNBP1 is widely expressed in brain, levels are reduced in brains of schizophrenia patients and a DTNBP1 polymorphism has been associated with reduced brain expression. Despite numerous genetic studies no DTNBP1 polymorphism has been strongly implicated in schizophrenia aetiology. Using a haplotype block-based gene-tagging approach we genotyped 13 SNPs in DTNBP1 to investigate possible associations with DTNBP1 and schizophrenia. Four polymorphisms were found to be significantly associated with schizophrenia. The strongest association was found with an A/C SNP in intron 7 (rs9370822). Homozygotes for the C allele of rs9370822 were more than two and a half times as likely to have schizophrenia compared to controls. The other polymorphisms showed much weaker association and are less likely to be biologically significant. These results suggest that DTNBP1 is a good candidate for schizophrenia risk and rs9370822 is either functionally important or in disequilibrium with a functional SNP, although our observations should be viewed with caution until they are independently replicated.

Evidence of multiwall carbon nanotube deformation caused by poly(3-hexylthiophene) adhesion

We show that when a soft polymer like Poly(3-hexyl-thiophene) wraps multiwall nanotubes by coiling around the main axis, a localized deformation of the nanotube structure is observed. High resolution transmission electron microscopy shows that radial compressions of about 4% can take place, and could possibly lead to larger interlayer distance between the nanotube inner walls and reduce the innermost nanotube radius. The mechanical stress due to the polymer presence was confirmed by Raman spectroscopic observation of a gradual upshift of the carbon nanotube G-band when the polymer content in the composites was progressively increased. Vibrational spectroscopy also indicates that charge transfer from the polymer to the nanotubes is responsible for a peak frequency relative downshift for high P3HT-content samples. Continuously acquired transmission electron microscopy images at rising temperature show the MWCNT elastic compression and relaxation due to polymer rearrangement on the nanotube surface.

Integrating a sustainable academic career around scholarly learning and teaching activities

To be scholarly in learning and teaching is rigorous academic work. It demands: currency and command of both discipline subject matter and educational theory; inquiring, methodical, and reflective approaches; the collection, evaluation and documentation of evidence of learning and teaching efficacy; and, optimally, entails participation in and communication among a community of teaching professionals. This chapter examines the author’s own practice in this regard to explicate the ‘how’ and ‘why’ of scholarly and scholarship approaches, as much as the ‘what’ and ‘where’ of that endeavour. In doing so, this meta‐analysis is made ‘community property’, in the same way that Shulman (1993: 6) exhorted we ‘change the status of teaching from private to community property’ so that teaching might be more greatly valued in the academy.

Assurance of learning : the role of work integrated learning and industry partners

In the partnering with students and industry it is important for universities to recognize and value the nature of knowledge and learning that emanates from work integrated learning experiences is different to formal university based learning. Learning is not a by-product of work rather learning is fundamental to engaging in work practice. Work integrated learning experiences provide unique opportunities for students to integrate theory and practice through the solving of real world problems. This paper reports findings to date of a project that sought to identify key issues and practices faced by academics, industry partners and students engaged in the provision and experience of work integrated learning within an undergraduate creative industries program at a major metropolitan university. In this paper, those findings are focused on some of the particular qualities and issues related to the assessment of learning at and through the work integrated experience. The findings suggest that the assessment strategies needed to better value the knowledges and practices of the Creative Industries. The paper also makes recommendations about how industry partners might best contribute to the assessment of students’ developing capabilities and to continuous reflection on courses and the assurance of learning agenda.

Uncovering the evidence of non-expert nephrology nursing practice

Expertise in nursing has been widely studied although there have been no previous studies into what constitutes expertise in nephrology (renal) nursing. This paper, which is abstracted from a larger study into the acquisition and exercise of nephrology nursing expertise, provides evidence of the characteristics and practices of non-expert nephrology nurses. Using the grounded theory method, the study took place in one renal unit in New South Wales, Australia, and involved six non-expert and 11 expert nurses. Sampling was purposive then theoretical. Simultaneous data collection and analysis using participant observation, review of nursing documentation and semistructured interviews was undertaken. The study revealed a three-stage skills-acquisitive process that was identified as non-expert, experienced non-expert and expert stages. Non-expert nurses showed superficial nephrology nursing knowledge and limited experience; they were acquiring basic nephrology nursing skills and possessed a narrow focus of practice.

Open problems in the security of learning

Machine learning has become a valuable tool for detecting and preventing malicious activity. However, as more applications employ machine learning techniques in adversarial decision-making situations, increasingly powerful attacks become possible against machine learning systems. In this paper, we present three broad research directions towards the end of developing truly secure learning. First, we suggest that finding bounds on adversarial influence is important to understand the limits of what an attacker can and cannot do to a learning system. Second, we investigate the value of adversarial capabilities-the success of an attack depends largely on what types of information and influence the attacker has. Finally, we propose directions in technologies for secure learning and suggest lines of investigation into secure techniques for learning in adversarial environments. We intend this paper to foster discussion about the security of machine learning, and we believe that the research directions we propose represent the most important directions to pursue in the quest for secure learning.