Weight vs magnetization enumerator for Gallager codes

We propose a method to determine the critical noise level for decoding Gallager type low density parity check error correcting codes. The method is based on the magnetization enumerator (¸M), rather than on the weight enumerator (¸W) presented recently in the information theory literature. The interpretation of our method is appealingly simple, and the relation between the different decoding schemes such as typical pairs decoding, MAP, and finite temperature decoding (MPM) becomes clear. Our results are more optimistic than those derived via the methods of information theory and are in excellent agreement with recent results from another statistical physics approach.

A note on DEA models in technology selection:an improvement of Karsak and Ahiska's approach

This paper introduces a compact form for the maximum value of the non-Archimedean in Data Envelopment Analysis (DEA) models applied for the technology selection, without the need to solve a linear programming (LP). Using this method the computational performance the common weight multi-criteria decision-making (MCDM) DEA model proposed by Karsak and Ahiska (International Journal of Production Research, 2005, 43(8), 1537-1554) is improved. This improvement is significant when computational issues and complexity analysis are a concern.

Investigation of potential intervention targets to improve insulin action: a therapeutic approach

Impaired insulin action (insulin resistance) is a key factor in the pathogenesis of diabetes mellitus. To investigate therapeutic targets against insulin resistance, this thesis explores the mechanism of action of pharmacological agents and exogenous peptides known or suspected to modify insulin action. These included leptin, a hormone primarily involved in the regulation of body weight; sibutramine, an antiobesity agent; plant-derived compounds (pinitol and chamaemeloside) and agents known to affect insulin sensitivity, e.g. metformin, tolbutamide, thiazolidinediones, vanadyl sulphate and thioctic acid. Models used for investigation included the L6 skeletal muscle cell line and isolated skeletal muscles. In vivo studies were undertaken to investigate glycaemia, insulinaemia, satiety and body weight in streptozotocin-induced diabetic mice and obese (ob/ob) mice. Leptin acutely altered insulin action in skeletal muscle cells via the short form of the leptin receptor. This direct action of leptin was mediated via a pathway involving PI 3-kinase but not Jak2. The active metabolites of sibutramine had antidiabetic properties in vivo and directly improved insulin sensitivity in vitro. This effect appeared to be conducted via a non-PI 3-kinase-mediated increase in protein synthesis with facilitated glucose transport, and was independent of the serotonin and noradrenaline reuptake inhibition produced by sibutramine. Pinitol (a methyl inositol) had an insulin mimetic effect and was an effective glucose-lowering agent in insulinopenic states, acting directly on skeletal muscle. Conversely chamaemeloside appeared to improve glucose tolerance without directly altering glucose transport. Metformin directly increased basal glucose uptake independently of PI 3-kinase, possibly via an increase in the intrinsic activity of glucose transporters. Neither tolbutamide nor thiazolidinediones directly altered insulin sensitivity in L6 skeletal muscle cells: however vanadyl sulphate and thioctic acid increased glucose transport but appeared to exert toxic effects at therapeutic concentrations. Examination of glucose transport in skeletal muscle in this thesis has identified various components of post-receptor insulin signalling pathways which may be targeted to ameliorate insulin resistance. Type 2 Diabetes Mellitus Obesity L6 Skeletal Muscle Cells Glucose Transport Insulin Signalling 2

A novel adaptive weight selection algorithm for multi-objective multi-agent reinforcement learning

To solve multi-objective problems, multiple reward signals are often scalarized into a single value and further processed using established single-objective problem solving techniques. While the field of multi-objective optimization has made many advances in applying scalarization techniques to obtain good solution trade-offs, the utility of applying these techniques in the multi-objective multi-agent learning domain has not yet been thoroughly investigated. Agents learn the value of their decisions by linearly scalarizing their reward signals at the local level, while acceptable system wide behaviour results. However, the non-linear relationship between weighting parameters of the scalarization function and the learned policy makes the discovery of system wide trade-offs time consuming. Our first contribution is a thorough analysis of well known scalarization schemes within the multi-objective multi-agent reinforcement learning setup. The analysed approaches intelligently explore the weight-space in order to find a wider range of system trade-offs. In our second contribution, we propose a novel adaptive weight algorithm which interacts with the underlying local multi-objective solvers and allows for a better coverage of the Pareto front. Our third contribution is the experimental validation of our approach by learning bi-objective policies in self-organising smart camera networks. We note that our algorithm (i) explores the objective space faster on many problem instances, (ii) obtained solutions that exhibit a larger hypervolume, while (iii) acquiring a greater spread in the objective space.

Description and preliminary results from a structured specialist behavioural weight management group intervention:Specialist Lifestyle Management (SLiM) programme

Background -  Specialist Lifestyle Management (SLiM) is a structured patient education and self-management group weight management programme. Each session is run monthly over a 6-month period providing a less intensive long-term approach. The groups are patient-centred incorporating educational, motivational, behavioural and cognitive elements. The theoretical background, programme structure and preliminary results of SLiM are presented. Subjects/methods - The study was a pragmatic service evaluation of obese patients with a body mass index (BMI) ≥35 kg/m2 with comorbidity or ≥40 kg/m2 without comorbidity referred to a specialist weight management service in the West Midlands, UK. 828 patients were enrolled within SLiM over a 48-month period. Trained facilitators delivered the programme. Preliminary anonymised data were analysed using the intention-to-treat principle. The primary outcome measure was weight loss at 3 and 6 months with comparisons between completers and non-completers performed. The last observation carried forward was used for missing data. Results - Of the 828 enrolled within SLiM, 464 completed the programme (56%). The mean baseline weight was 135 kg (BMI=49.1 kg/m2) with 87.2% of patients having a BMI≥40 kg/m2 and 12.4% with BMI≥60 kg/m2. The mean weight change of all patients enrolled was −4.1 kg (95% CI −3.6 to −4.6 kg, p=0.0001) at the end of SLiM, with completers (n=464) achieving −5.5 kg (95% CI −4.2 to −6.2 kg, p=0.0001) and non-completers achieving −2.3 kg (p=0.0001). The majority (78.6%) who attended the 6-month programme achieved weight loss with 32.3% achieving a ≥5% weight loss. Conclusions - The SLiM programme is an effective group intervention for the management of severe and complex obesity.