Adaptive uncertain information fusion to enhance plan selection in BDI agent systems

Correctly modelling and reasoning with uncertain information from heterogeneous sources in large-scale systems is critical when the reliability is unknown and we still want to derive adequate conclusions. To this end, context-dependent merging strategies have been proposed in the literature. In this paper we investigate how one such context-dependent merging strategy (originally defined for possibility theory), called largely partially maximal consistent subsets (LPMCS), can be adapted to Dempster-Shafer (DS) theory. We identify those measures for the degree of uncertainty and internal conflict that are available in DS theory and show how they can be used for guiding LPMCS merging. A simplified real-world power distribution scenario illustrates our framework. We also briefly discuss how our approach can be incorporated into a multi-agent programming language, thus leading to better plan selection and decision making.

Proactive Relay Selection with Joint Impact of Hardware Impairment and Co-channel Interference

In this paper, we investigate the end-to-end performance of dual-hop proactive decode-and-forward relaying networks with Nth best relay selection in the presence of two practical deleterious effects: i) hardware impairment and ii) cochannel interference. In particular, we derive new exact and asymptotic closed-form expressions for the outage probability and average channel capacity of Nth best partial and opportunistic relay selection schemes over Rayleigh fading channels. Insightful discussions are provided. It is shown that, when the system cannot select the best relay for cooperation, the partial relay selection scheme outperforms the opportunistic method under the impact of the same co-channel interference (CCI). In addition, without CCI but under the effect of hardware impairment, it is shown that both selection strategies have the same asymptotic channel capacity. Monte Carlo simulations are presented to corroborate our analysis.

Transmit Antenna Selection in Cognitive MIMO Relaying with Multiple Primary Transceivers

We consider transmit antenna selection (TAS) in cognitive multiple-input multiple-output (MIMO) relay networks, as an interference-aware design for secondary users (SUs) to ensure power and interference constraints of multiple primary users (PUs). In doing so, we derive new exact and asymptotic expressions for the outage probability of TAS with maximal ratio combining (TAS/MRC) and with selection combining (TAS/SC) over Rayleigh fading. The proposed analysis and simulations highlight that TAS/MRC and TAS/SC with decode-and-forward relaying achieve the same diversity order in cognitive MIMO networks, which scales with the minimum number of antennas at the SUs. Furthermore, we accurately characterize the outage gap between TAS/MRC and TAS/SC relaying as a concise ratio of their array gains.

Relay Selection for Security Enhancement in Cognitive Relay Networks

This letter proposes several relay selection policies for secure communication in cognitive decode-and-forward (DF) relay networks, where a pair of cognitive relays are opportunistically selected for security protection against eavesdropping. The first relay transmits the secrecy information to the destination,
and the second relay, as a friendly jammer, transmits the jamming signal to confound the eavesdropper. We present new exact closed-form expressions for the secrecy outage probability. Our analysis and simulation results strongly support our conclusion that the proposed relay selection policies can enhance the performance of secure cognitive radio. We also confirm that the error floor phenomenon is created in the absence of jamming.

Generalized Selection Combining for Cognitive Relay Networks over Nakagami-m Fading

We consider transmit antenna selection with receive generalized selection combining (TAS/GSC) for cognitive decodeand-forward (DF) relaying in Nakagami-m fading channels. In an effort to assess the performance, the probability density function and the cumulative distribution function of the endto-end SNR are derived using the moment generating function, from which new exact closed-form expressions for the outage probability and the symbol error rate are derived. We then derive a new closed-form expression for the ergodic capacity. More importantly, by deriving the asymptotic expressions for the outage probability and the symbol error rate, as well as the high SNR approximations of the ergodic capacity, we establish new design insights under the two distinct constraint scenarios: 1) proportional interference power constraint, and 2) fixed interference power constraint. Several pivotal conclusions are reached. For the first scenario, the full diversity order of the
outage probability and the symbol error rate is achieved, and the high SNR slope of the ergodic capacity is 1/2. For the second scenario, the diversity order of the outage probability and the symbol error rate is zero with error floors, and the high SNR slope of the ergodic capacity is zero with capacity ceiling.

Cross-participant modelling based on joint or disjoint feature selection: an fMRI conceptual decoding study

Multivariate classification techniques have proven to be powerful tools for distinguishing experimental conditions in single sessions of functional magnetic resonance imaging (fMRI) data. But they are vulnerable to a considerable penalty in classification accuracy when applied across sessions or participants, calling into question the degree to which fine-grained encodings are shared across subjects. Here, we introduce joint learning techniques, where feature selection is carried out using a held-out subset of a target dataset, before training a linear classifier on a source dataset. Single trials of functional MRI data from a covert property generation task are classified with regularized regression techniques to predict the semantic class of stimuli. With our selection techniques (joint ranking feature selection (JRFS) and disjoint feature selection (DJFS)), classification performance during cross-session prediction improved greatly, relative to feature selection on the source session data only. Compared with JRFS, DJFS showed significant improvements for cross-participant classification. And when using a groupwise training, DJFS approached the accuracies seen for prediction across different sessions from the same participant. Comparing several feature selection strategies, we found that a simple univariate ANOVA selection technique or a minimal searchlight (one voxel in size) is appropriate, compared with larger searchlights.

Plan Selection for Probabilistic BDI Agents

When an agent wants to fulfill its desires about the world, the agent usually has multiple plans to choose from and these plans have different pre-conditions and additional effects in addition to achieving its goals. Therefore, for further reasoning and interaction with the world, a plan selection strategy (usually based on plan cost estimation) is mandatory for an autonomous agent. This demand becomes even more critical when uncertainty on the observation of the world is taken into account, since in this case, we consider not only the costs of different plans, but also their chances of success estimated according to the agent's beliefs. In addition, when multiple goals are considered together, different plans achieving the goals can be conflicting on their preconditions (contexts) or the required resources. Hence a plan selection strategy should be able to choose a subset of plans that fulfills the maximum number of goals while maintaining context consistency and resource-tolerance among the chosen plans. To address the above two issues, in this paper we first propose several principles that a plan selection strategy should satisfy, and then we present selection strategies that stem from the principles, depending on whether a plan cost is taken into account. In addition, we also show that our selection strategy can partially recover intention revision.

Aspirin and folic acid for the prevention of recurrent colorectal adenomas

BACKGROUND & AIMS: Although observational studies have found regular aspirin use to be associated with a reduced risk of colorectal neoplasia, results from randomized trials using aspirin have been inconsistent. Dietary folate intake also has been found to be associated with a reduced risk of colorectal neoplasms in observational studies.

METHODS: A multicenter, randomized, double-blind trial of aspirin (300 mg/day) and folate supplements (0.5 mg/day) to prevent colorectal adenoma recurrence was performed using a 2 x 2 factorial design. All patients had an adenoma (>/=0.5 cm) removed in the 6 months before recruitment and were followed-up at 4-month intervals with a second colonoscopy after approximately 3 years. The primary outcome measure was a colorectal adenoma diagnosed after baseline.

RESULTS: A total of 945 patients were recruited into the study, of whom 853 (90.3%) underwent a second colonoscopy. In total, 99 (22.8%) of 434 patients receiving aspirin had a recurrent adenoma compared with 121 (28.9%) of 419 patients receiving placebo (relative risk, 0.79; 95% confidence interval [CI], 0.63-0.99). A total of 104 patients developed an advanced colorectal adenoma; 41 (9.4%) of these were in the aspirin group and 63 (15.0%) were in the placebo group (relative risk, 0.63; 95% CI, 0.43-0.91). Folate supplementation was found to have no effect on adenoma recurrence (relative risk, 1.07; 95% CI, 0.85-1.34).

CONCLUSIONS: Aspirin (300 mg/day) but not folate (0.5 mg/day) use was found to reduce the risk of colorectal adenoma recurrence, with evidence that aspirin could have a significant role in preventing the development of advanced lesions.

Distinguishing Species of Geoemyclid and Trionychid Turtles from Shell Fragments:Evidence from the Pleistocene at Niah Caves, Sarawak

Fragments of chelonian carapace and plastral dermal plates are well-represented from archaeological sites in the world's dry and wet tropics. However, although these bones are easily identified at an element level, few archaeological reports have explored the potential of using features of form and surface sculpturing as a way to refine that identification to genus or species. The ability to achieve such a refinement would benefit environmental and human subsistence strategy models alike. The objective of the current paper was to isolate recurrent and readily visible surface characteristics on the dermal plates from a selection of commonly occurring Southeast Asian hard- and soft-shelled turtles. Using these criteria, analysis is made of the chelonian assemblage from pre- and post-Last Glacial Maximum (LGM) cultural deposits in the West Mouth of Niah Cave. Copyright (C) 2009 John Wiley & Sons, Ltd.