Davide Morelli

Experience-Driven Procedural Music Generation for Games

As video games have grown from crude and simple circuit-based artefacts to a multibillion dollar worldwide industry, video-game music has become increasingly adaptive. Composers have had to use new techniques to avoid the traditional, event-based approach where music is composed mostly of looped audio tracks, which can lead to music that is too repetitive. In addition, these cannot scale well in the design of todays games, which have become increasingly complex and nonlinear in narrative. This paper outlines the use of experience-driven procedural music generation, to outline possible ways forward in the dynamic generation of music and audio according to user gameplay metrics.

The Association between Work-Related Rumination and Heart Rate Variability: A Field Study

The objective of this study was to examine the association between perseverative cognition in the form of work-related rumination, and heart rate variability (HRV). We tested the hypothesis that high ruminators would show lower vagally mediated HRV relative to low ruminators during their leisure time. Individuals were classified as being low (n = 17) or high ruminators (n = 19), using the affective scale on the work-related rumination measure. HRV was assessed using a wrist sensor band (Microsoft Band 2). HRV was sampled between 8 pm and 10 pm over three workday evenings (Monday to Wednesday) while individuals carried out their normal evening routines. Compared to the low ruminators, high affective ruminators demonstrated lower HRV in the form of root mean square successive differences (RMSSDs), relative to the low ruminators, indicating lower parasympathetic activity. There was no significant difference in heart rate, or activity levels between the two groups during the recording periods. The current findings of this study may have implications for the design and delivery of interventions to help individuals unwind post work and to manage stress more effectively. Limitations and implications for future research are discussed.

DropIn: Making reservoir computing neural networks robust to missing inputs by dropout

The paper presents a novel, principled approach to train recurrent neural networks from the Reservoir Computing family that are robust to missing part of the input features at prediction time. By building on the ensembling properties of Dropout regularization, we propose a methodology, named DropIn, which efficiently trains a neural model as a committee machine of subnetworks, each capable of predicting with a subset of the original input features. We discuss the application of the DropIn methodology in the context of Reservoir Computing models and targeting applications characterized by input sources that are unreliable or prone to be disconnected, such as in pervasive wireless sensor networks and ambient intelligence. We provide an experimental assessment using real-world data from such application domains, showing how the Dropin methodology allows to maintain predictive performances comparable to those of a model without missing features, even when 20%–50% of the inputs are not available.

A compositional model to characterize software and hardware from their resource usage

Since the introduction of laptops and mobile devices, there has been a strong research focus towards the energy efficiency of hardware. Many papers, both from academia and industrial research labs, focus on methods and ideas to lower power consumption in order to lengthen the battery life of portable device components. Much less effort has been spent on defining the responsibility of software in the overall computational system’s energy consumption. Some attempts have been made to describe the energy behaviour of software, but none of them abstract from the physical machine where the measurements were taken. In our opinion this is a strong drawback because results can not be generalized. We propose a measuring method and a set of algebraic tools that can be applied to resource usage measurements. These tools are expressive and show insights on how the hardware consumes energy (or other resources), but are equally able to describe how efficiently the software exploits hardware characteristics. The method is based on the idea of decomposing arbitrary programs into linear combinations of benchmarks of a test-bed without the need to analyse a program’s source code by employing a black box approach, measuring only its resource usage.

Information processing at work: On energy-aware algorithm design

It is common experience to upgrade firmware of mobile devices and obtain longer battery life, living proof of how software affects power consumption of a device. Despite this empirical observation, there is a lack for models and methodologies correlating computations with power consumption. In this paper we propose a methodology for conducting measures which result independent of the underlying system running the algorithm/software to be tested. Early experimental results are presented and discussed, showing that this methodology is robust and can be used in many settings. We thus adopt it to study the impact of computation and pattern of memory accesses onto the energy-profile of an algorithm when executed on different processors and architectures, thus achieving some surprising insights on green algorithm design.

A high-level and accurate energy model of parallel and concurrent workloads

The ability to predict the energy needed by a system to perform a task, or several concurrent parallel tasks, allows the scheduler to enforce energy‐aware policies while providing acceptable performance. The approaches in literature to model energy consumption of tasks usually focus on low‐level descriptors and require invasive instrumentation of the computational environment. We developed an energy model and a methodology to automatically extract features that characterize the computational environment relying only on a single power meter that measures the energy consumption of the whole system. Once the model has been built, the energy consumption of concurrent tasks can be calculated, with a statistically insignificant error, even without any power meter. We show that our model can predict with high accuracy, even only using the utilization time of the cores in a high‐performance computing enclosure, without using performance counters. Hence, the model could be easily applicable to heterogeneous systems, where collecting representative performance counters can be problematic. Copyright

The application of a biofeedback breathing app as an intervention to augment post-stress physiological recovery (Preprint)

Background The speed of physiological recovery from stress may be a marker for cardiovascular disease risk. Stress management programs that incorporate guided breathing have been shown to moderate the stress response and augment recovery. Objective The aim of this study was to examine the effectiveness of an app-based brief relaxation intervention (BioBase) for facilitating physiological recovery in individuals exposed to a brief psychological stressor. Methods A total of 75 participants (44 women) completed a stressor speech task and were randomly assigned to one of three conditions: control, rumination, or an app-based relaxation breathing (BioBase) conditions. Heart rate variability (HRV) was assessed as a measure of autonomic function at baseline (6 min), during stress (6 min), and during recovery (6 min). Results There was a significant increase in subjective stress following stress exposure, but the ratings returned to baseline after recovery in all three groups. In addition, there was a significant decrease in vagally mediated HRV in the poststress period. During recovery, the root mean square of successive differences (P<.001), the percentage of successive interbeat (RR) intervals that differ by >50 ms (pNN50; P<.001), and high-frequency (P<.02) HRV were significantly higher in the BioBase breathing condition than the rumination and control conditions. There was no difference in HRV values between the rumination and control conditions during recovery. Conclusions App-based relaxed breathing interventions could be effective in reducing cardiovascular disease risk. These results provide additional utility of biofeedback breathing in augmenting physiological recovery from psychological stress.

Profiling the Propagation of Error from PPG to HRV Features in a WearablePhysiological-Monitoring Device

Wearable physiological monitors are becoming increasingly commonplace in the consumer domain, but in literature there exists no substantive studies of their performance when measuring the physiology of ambulatory patients. In this Letter, the authors investigate the reliability of the heart-rate (HR) sensor in an exemplar ‘wearable’ wrist-worn monitoring system (the Microsoft Band 2); their experiments quantify the propagation of error from (i) the photoplethysmogram (PPG) acquired by pulse oximetry, to (ii) estimation of HR, and (iii) subsequent calculation of HR variability (HRV) features. Their experiments confirm that motion artefacts account for the majority of this error, and show that the unreliable portions of HR data can be removed, using the accelerometer sensor from the wearable device. The experiments further show that acquired signals contain noise with substantial energy in the high-frequency band, and that this contributes to subsequent variability in standard HRV features often used in clinical practice. The authors finally show that the conventional use of long-duration windows of data is not needed to perform accurate estimation of time-domain HRV features.

Accurate Blind Predictions of OpenFOAM Energy Consumption Using the LBM Prediction Model

The ability to predict the energy consumption of an HPC task, varying the number of assigned nodes, can lead to the ability to assign the correct number of nodes to tasks, saving large amount of energy.