On Variable Selection of a Nonlinear Non-parametric System with a Limited Data Set: A Stepwise Algorithm

his paper considers a problem of identification for a high dimensional nonlinear non-parametric system when only a limited data set is available. The algorithms are proposed for this purpose which exploit the relationship between the input variables and the output and further the inter-dependence of input variables so that the importance of the input variables can be established. A key to these algorithms is the non-parametric two stage input selection algorithm.

Testing the channel-based model of duration perception

The channel-based model of duration perception postulates the existence of neural mechanisms that respond selectively to a narrow range of stimulus durations centred on their preferred duration (Heron et al Proceedings of the Royal Society B 279 690–698). In principle the channel-based model could
explain recent reports of adaptation-induced, visual duration compression effects (Johnston et al Current Biology 16 472–479; Curran and Benton Cognition 122 252–257); from this perspective duration compression is a consequence of the adapting stimuli being presented for a longer duration than the test stimuli. In the current experiment observers adapted to a sequence of moving random dot patterns at the same retinal position, each 340ms in duration and separated by a variable (500–1000ms) interval. Following adaptation observers judged the duration of a 600ms test stimulus at the same location. The test stimulus moved in the same, or opposite, direction as the adaptor. Contrary to the channel-based
model’s prediction, test stimulus duration appeared compressed, rather than expanded, when it moved in the same direction as the adaptor. That test stimulus duration was not distorted when moving in the opposite direction further suggests that visual timing mechanisms are influenced by additional neural processing associated with the stimulus being timed.

The timing and spatiotemporal patterning of Neanderthal disappearance

ThetimingofNeanderthal disappearanceandtheextent to whichthey overlapped with the earliest incoming anatomically modern humans (AMHs)inEurasia arekey questions inpalaeoanthropology1,2 .Deter- mining the spatiotemporal relationship between the two populations is crucial if we are to understand the processes, timing and reasons leading to the disappearance of Neanderthals and the likelihood of cultural and genetic exchange. Serious technical challenges, however, havehinderedreliable datingof the period,as theradiocarbonmethod reaches its limit at 50,000 years ago3 .Herewe apply improved accel- erator mass spectrometry 14C techniques to construct robust chro- nologies from 40 key Mousterian and Neanderthal archaeological sites, ranging fromRussia toSpain.Bayesianagemodellingwas used togenerate probability distributionfunctions todetermine the latest appearancedate.Weshowthat theMousterianendedby41,030–39,260 calibratedyears BP(at95.4%probability) acrossEurope.Wealsodem- onstrate that succeeding ‘transitional’ archaeological industries, one ofwhich has beenlinked withNeanderthals (Cha ˆtelperronian)4 ,end at a similar time. Our data indicate that the disappearance of Nean- derthals occurred at different times in different regions.Comparing the data with results obtained fromthe earliest datedAMHsites in Europe, associated with the Uluzzian technocomplex5 , allows us to quantify the temporal overlap between the two human groups. The results revealasignificantoverlap of 2,600–5,400years (at 95.4%prob- ability).This hasimportant implications formodels seeking toexplain the cultural, technological and biological elements involved in the replacement of Neanderthals byAMHs.Amosaic of populations in Europe during the Middle to Upper Palaeolithic transition suggests that there was ample time for the transmission of cultural and sym- bolic behaviours, as well as possible genetic exchanges, between the two groups.

Exploiting dynamic timing margins in microprocessors for frequency-over-scaling with instruction-based clock adjustment

Static timing analysis provides the basis for setting the clock period of a microprocessor core, based on its worst-case critical path. However, depending on the design, this critical path is not always excited and therefore dynamic timing margins exist that can theoretically be exploited for the benefit of better speed or lower power consumption (through voltage scaling). This paper introduces predictive instruction-based dynamic clock adjustment as a technique to trim dynamic timing margins in pipelined microprocessors. To this end, we exploit the different timing requirements for individual instructions during the dynamically varying program execution flow without the need for complex circuit-level measures to detect and correct timing violations. We provide a design flow to extract the dynamic timing information for the design using post-layout dynamic timing analysis and we integrate the results into a custom cycle-accurate simulator. This simulator allows annotation of individual instructions with their impact on timing (in each pipeline stage) and rapidly derives the overall code execution time for complex benchmarks. The design methodology is illustrated at the microarchitecture level, demonstrating the performance and power gains possible on a 6-stage OpenRISC in-order general purpose processor core in a 28nm CMOS technology. We show that employing instruction-dependent dynamic clock adjustment leads on average to an increase in operating speed by 38% or to a reduction in power consumption by 24%, compared to traditional synchronous clocking, which at all times has to respect the worst-case timing identified through static timing analysis.

Timing of response differentiation in human motor cortex during a speeded Go/No-go task

We explored the brain's ability to quickly prevent a pre-potent but unwanted motor response. To address this, transcranial magnetic stimulation was delivered over the motor cortex (hand representation) to probe excitability changes immediately after somatosensory cues prompted subjects to either move as fast as possible or withhold movement. Our results showed a difference in motor cortical excitability 90 ms post-stimulus contingent on cues to either promote or prevent movement. We suggest that our study design emphasizing response speed coupled with well-defined early probes allowed us to extend upon similar past investigations into the timing of response inhibition.

Investigating the Implications of a Variable RBE on Proton Dose Fractionation Across a Clinical Pencil Beam Scanned Spread-Out Bragg Peak

Purpose: To investigate the clinical implications of a variable relative biological effectiveness (RBE) on proton dose fractionation. Using acute exposures, the current clinical adoption of a generic, constant cell killing RBE has been shown to underestimate the effect of the sharp increase in linear energy transfer (LET) in the distal regions of the spread-out Bragg peak (SOBP). However, experimental data for the impact of dose fractionation in such scenarios are still limited.

Methods and Materials: Human fibroblasts (AG01522) at 4 key depth positions on a clinical SOBP of maximum energy 219.65 MeV were subjected to various fractionation regimens with an interfraction period of 24 hours at Proton Therapy Center in Prague, Czech Republic. Cell killing RBE variations were measured using standard clonogenic assays and were further validated using Monte Carlo simulations and parameterized using a linear quadratic formalism.

Results: Significant variations in the cell killing RBE for fractionated exposures along the proton dose profile were observed. RBE increased sharply toward the distal position, corresponding to a reduction in cell sparing effectiveness of fractionated proton exposures at higher LET. The effect was more pronounced at smaller doses per fraction. Experimental survival fractions were adequately predicted using a linear quadratic formalism assuming full repair between fractions. Data were also used to validate a parameterized variable RBE model based on linear α parameter response with LET that showed considerable deviations from clinically predicted isoeffective fractionation regimens.

Conclusions: The RBE-weighted absorbed dose calculated using the clinically adopted generic RBE of 1.1 significantly underestimates the biological effective dose from variable RBE, particularly in fractionation regimens with low doses per fraction. Coupled with an increase in effective range in fractionated exposures, our study provides an RBE dataset that can be used by the modeling community for the optimization of fractionated proton therapy.

Towards the 2021 UK Census imputation strategy: Response mode as a matching variable in a donor-based approach?

Since July 2014, the Office for National Statistics has committed to a predominantly online 2021 UK Census. Item-level imputation will play an important role in adjusting the 2021 Census database. Research indicates that the internet may yield cleaner data than paper based capture and attract people with particular characteristics. Here, we provide preliminary results from research directed at understanding how we might manage these features in a 2021 UK Census imputation strategy. Our findings suggest that if using a donor-based imputation method, it may need to consider including response mode as a matching variable in the underlying imputation model.

Variable Window Power Spectral Density Attack

Side channel attacks permit the recovery of the secret key held within a cryptographic device. This paper presents a new EM attack in the frequency domain, using a power spectral density analysis that permits the use of variable spectral window widths for each trace of the data set and demonstrates how this attack can therefore overcome both inter-and intra-round random insertion type countermeasures. We also propose a novel re-alignment method exploiting the minimal power markers exhibited by electromagnetic emanations. The technique can be used for the extraction and re-alignment of round data in the time domain.

Replicability of data collected for empirical estimation of relative pollen productivity

The effects of repeated survey and fieldwork timing on data derived from a recently proposed standard field methodology for empirical estimation of Relative Pollen Productivity have been tested. Seasonal variations in vegetation and associated pollen assemblages were studied in three contrasting cultural habitat types; semi-natural ancient woodlands, lowland heaths, and unimproved, traditionally managed hay meadows. Results show that in woodlands and heathlands the standard method generates vegetation data with a reasonable degree of similarity throughout the field season, though in some instances additional recording of woodland canopy cover should be undertaken, and differences were greater for woodland understorey taxa than for arboreal taxa. Large differences in vegetation cover were observed over the field season in the grassland community, and matching the phenological timing of surveys within and between studies is clearly important if RPP estimates from these sites are to be comparable. Pollen assemblages from closely co-located moss polsters collected on different visits are shown to be variable in all communities, to a greater degree than can be explained by the sampling error associated with pollen counting, and further study of moss polsters as pollen traps is recommended.

Architecture for real-time coordination of multiple autonomous mobile units

Interest on using teams of mobile robots has been growing, due to their potential to cooperate for diverse purposes, such as rescue, de-mining, surveillance or even games such as robotic soccer. These applications require a real-time middleware and wireless communication protocol that can support an efficient and timely fusion of the perception data from different robots as well as the development of coordinated behaviours. Coordinating several autonomous robots towards achieving a common goal is currently a topic of high interest, which can be found in many application domains. Despite these different application domains, the technical problem of building an infrastructure to support the integration of the distributed perception and subsequent coordinated action is similar. This problem becomes tougher with stronger system dynamics, e.g., when the robots move faster or interact with fast objects, leading to tighter real-time constraints. This thesis work addressed computing architectures and wireless communication protocols to support efficient information sharing and coordination strategies taking into account the real-time nature of robot activities. The thesis makes two main claims. Firstly, we claim that despite the use of a wireless communication protocol that includes arbitration mechanisms, the self-organization of the team communications in a dynamic round that also accounts for variable team membership, effectively reduces collisions within the team, independently of its current composition, significantly improving the quality of the communications. We will validate this claim in terms of packet losses and communication latency. We show how such self-organization of the communications can be achieved in an efficient way with the Reconfigurable and Adaptive TDMA protocol. Secondly, we claim that the development of distributed perception, cooperation and coordinated action for teams of mobile robots can be simplified by using a shared memory middleware that replicates in each cooperating robot all necessary remote data, the Real-Time Database (RTDB) middleware. These remote data copies, which are updated in the background by the selforganizing communications protocol, are extended with age information automatically computed by the middleware and are locally accessible through fast primitives. We validate our claim showing a parsimonious use of the communication medium, improved timing information with respect to the shared data and the simplicity of use and effectiveness of the proposed middleware shown in several use cases, reinforced with a reasonable impact in the Middle Size League of RoboCup.

A Spiking Neural Network Model of the Medial Superior Olive using Spike Timing Dependent Plasticity for Sound Localisation

Sound localization can be defined as the ability to identify the position of an input sound source and is considered a powerful aspect of mammalian perception. For low frequency sounds, i.e., in the range 270 Hz-1.5 KHz, the mammalian auditory pathway achieves this by extracting the Interaural Time Difference between sound signals being received by the left and right ear. This processing is performed in a region of the brain known as the Medial Superior Olive (MSO). This paper presents a Spiking Neural Network (SNN) based model of the MSO. The network model is trained using the Spike Timing Dependent Plasticity learning rule using experimentally observed Head Related Transfer Function data in an adult domestic cat. The results presented demonstrate how the proposed SNN model is able to perform sound localization with an accuracy of 91.82% when an error tolerance of +/-10 degrees is used. For angular resolutions down to 2.5 degrees , it will be demonstrated how software based simulations of the model incur significant computation times. The paper thus also addresses preliminary implementation on a Field Programmable Gate Array based hardware platform to accelerate system performance.