Showing papers by "INESC-ID published in 2021"

A multi-country test of brief reappraisal interventions on emotions during the COVID-19 pandemic.

Abstract: The COVID-19 pandemic has increased negative emotions and decreased positive emotions globally. Left unchecked, these emotional changes might have a wide array of adverse impacts. To reduce negative emotions and increase positive emotions, we tested the effectiveness of reappraisal, an emotion-regulation strategy that modifies how one thinks about a situation. Participants from 87 countries and regions (n = 21,644) were randomly assigned to one of two brief reappraisal interventions (reconstrual or repurposing) or one of two control conditions (active or passive). Results revealed that both reappraisal interventions (vesus both control conditions) consistently reduced negative emotions and increased positive emotions across different measures. Reconstrual and repurposing interventions had similar effects. Importantly, planned exploratory analyses indicated that reappraisal interventions did not reduce intentions to practice preventive health behaviours. The findings demonstrate the viability of creating scalable, low-cost interventions for use around the world.

Kauri: Scalable BFT Consensus with Pipelined Tree-Based Dissemination and Aggregation

Abstract: With the growing commercial interest in blockchains, permissioned implementations have received increasing attention. Unfortunately, the BFT consensus algorithms that are the backbone of most of these blockchains scale poorly and offer limited throughput. Many state-of-the-art algorithms require a single leader process to receive and validate votes from a quorum of processes and then broadcast the result, which is inherently non-scalable. Recent approaches avoid this bottleneck by using dissemination/aggregation trees to propagate values and collect and validate votes. However, the use of trees increases the round latency, which ultimately limits the throughput for deeper trees. In this paper we propose Kauri, a BFT communication abstraction that can sustain high throughput as the system size grows, leveraging a novel pipelining technique to perform scalable dissemination and aggregation on trees. Our evaluation shows that Kauri outperforms the throughput of state-of-the-art permissioned blockchain protocols, such as HotStuff, by up to 28x. Interestingly, in many scenarios, the parallelization provided by Kauri can also decrease the latency.

Polygon Detection from a Set of Lines

Abstract: Detecting polygons defined by a set of line segments in a plane is an important step in the analysis of vectorial drawings. This paper presents an approach that combines several algorithms to detect basic polygons from a set of arbitrary line segments. The resulting algorithm runs in polynomial time and space, with complexities of O((N +M)) and O((N +M)) respectively, where N is the number of line segments and M is the number of intersections between line segments. Our choice of algorithms was made to strike a good compromise between efficiency and ease of implementation. The result is a simple and efficient solution to detect polygons from lines.

Dual Output and High Voltage Gain DC-DC Converter for PV and Fuel Cell Generators Connected to DC Bipolar Microgrids

Abstract: This paper introduces a new topology for a DC-DC converter with bipolar output and high voltage gain. The topology was designed with the aim to use only one active power switch. Besides the bipolar multiport output and high voltage gain this converter has another important feature, namely, it has a continuous input current. Due to the self-balancing bipolar outputs, the proposed topology is suitable for bipolar DC microgrids. Indeed, the topology balancing capability can achieve the two symmetrical voltage poles of bipolar DC microgrids. Furthermore, it is possible to create a midpoint in the output of the converter that can be directly connected to the ground of the DC power supply, avoiding common-mode leakage currents in critical applications such as transformerless grid-connect PV systems. The operating principle of the proposed topology will be supported by mathematical analysis. To validate and verify the characteristics of the presented topology, several experimental results are shown.

Humor–Robot Interaction: A Scoping Review of the Literature and Future Directions

Abstract: Humor is a pervasive feature of everyday social interactions that might be leveraged to improve human–robot interactions (HRI). Our goal is to evaluate how the use of humor can improve HRI and enhance the user’s perception of the robot, as well as to derive implications for future research and development of humorous robots. We conducted a systematic search of 7 digital libraries relevant in the areas of HRI and Psychology for papers that were relevant to our goal. We identified 431 records, published between 2000 and August of 2020, of which 12 matched our eligibility criteria. The included studies reported the results of original empirical research that involved direct or video-mediated interaction of humans and robots. Humor seems to have a positive effect in improving the user’s perception of the robot, as well as the user’s evaluation of the interaction. However, the included studies present a number of limitations in their approaches to robotic humor that need to be surpassed before reaching a final verdict on the value of humor in HRI.

From warm to hot starts: leveraging runtimes for the serverless era

Abstract: The serverless computing model leverages high-level languages, such as JavaScript and Java, to raise the level of abstraction for cloud programming. However, today's design of serverless computing platforms based on stateless short-lived functions leads to missed opportunities for modern runtimes to optimize serverless functions through techniques such as JIT compilation and code profiling. In this paper, we show that modern serverless platforms, such as AWS Lambda, do not fully leverage language runtime optimizations. We find that a significant number of function invocations running on warm containers are executed with unoptimized code (warm-starts), leading to orders of magnitude performance slowdowns. We explore the idea of exploiting the runtime knowledge spread throughout potentially thousands of nodes to profile and optimize code. To that end, we propose Ignite, a serverless platform that orchestrates runtimes across machines to run optimized code from the start (hot-start). We present evidence that runtime orchestration has the potential to greatly reduce cost and latency of serverless workloads by running optimized code across thousands of serverless functions.

SOAR: A Synthesis Approach for Data Science API Refactoring

Abstract: With the growth of the open-source data science community, both the number of data science libraries and the number of versions for the same library are increasing rapidly. To match the evolving APIs from those libraries, open-source organizations often have to exert manual effort to refactor the APIs used in the code base. Moreover, due to the abundance of similar open-source libraries, data scientists working on a certain application may have an abundance of libraries to choose, maintain and migrate between. The manual refactoring between APIs is a tedious and error-prone task. Although recent research efforts were made on performing automatic API refactoring between different languages, previous work relies on statistical learning with collected pairwise training data for the API matching and migration. Using large statistical data for refactoring is not ideal because such training data will not be available for a new library or a new version of the same library. We introduce Synthesis for Open-Source API Refactoring (SOAR), a novel technique that requires no training data to achieve API migration and refactoring. SOAR relies only on the documentation that is readily available at the release of the library to learn API representations and mapping between libraries. Using program synthesis, SOAR automatically computes the correct configuration of arguments to the APIs and any glue code required to invoke those APIs. SOAR also uses the interpreter's error messages when running refactored code to generate logical constraints that can be used to prune the search space. Our empirical evaluation shows that SOAR can successfully refactor 80% of our benchmarks corresponding to deep learning models with up to 44 layers with an average run time of 97.23 seconds, and 90% of the data wrangling benchmarks with an average run time of 17.31 seconds.

Distance-based phylogenetic inference from typing data: a unifying view.

Abstract: Typing methods are widely used in the surveillance of infectious diseases, outbreaks investigation and studies of the natural history of an infection. Moreover, their use is becoming standard, in particular with the introduction of high-throughput sequencing. On the other hand, the data being generated are massive and many algorithms have been proposed for a phylogenetic analysis of typing data, addressing both correctness and scalability issues. Most of the distance-based algorithms for inferring phylogenetic trees follow the closest pair joining scheme. This is one of the approaches used in hierarchical clustering. Moreover, although phylogenetic inference algorithms may seem rather different, the main difference among them resides on how one defines cluster proximity and on which optimization criterion is used. Both cluster proximity and optimization criteria rely often on a model of evolution. In this work, we review, and we provide a unified view of these algorithms. This is an important step not only to better understand such algorithms but also to identify possible computational bottlenecks and improvements, important to deal with large data sets.

Ad Hoc Teamwork in the Presence of Non-stationary Teammates.

Abstract: In this paper we address the problem of ad hoc teamwork and contribute a novel approach, PPAS, that is able to handle non-stationary teammates. Current approaches to ad hoc teamwork assume that the (potentially unknown) teammates behave in a stationary way, which is a significant limitation in real world conditions, since humans and other intelligent systems do not necessarily follow strict policies. In our work we highlight the current limitations of state-of-the-art approaches to ad hoc teamwork problem in the presence of non-stationary teammate, and propose a novel solution that alleviates the stationarity assumption by combining ad hoc teamwork with adversarial online prediction. The proposed architecture is called PLASTIC Policy with Adversarial Selection, or PPAS. We showcase the effectiveness of our approach through an empirical evaluation in the half-field offense environment. Our results show that it is possible to cooperate in an ad hoc manner with non-stationary teammates in complex environments.

Prosociality in Cyberspace: Developing Emotion and Behavioral Regulation to Decrease Aggressive Communication

Abstract: Different forms of verbal aggression are often present in cyberbullying, which may impair executive function skills that enable the regulation of emotions and behavior. Emotion and behavioral regulation has been associated with better social adjustment and more positive interactions between peers. This study aimed to understand if fostering emotion and behavioral regulation strategies could decrease aggressive communication. A quasi-experimental longitudinal design, based on a Twitter client mobile application, with pre-posttest measures was used. For the application, we explored different machine learning approaches, including computational intelligence methods. Multilevel linear modeling and frequency analyses were performed. A convenience sample of 218 adolescents (Mage = 14.67, SD = 0.84, 53% female) participated in the study. Results suggest that a Twitter client mobile application intervention based on emotion and behavioral regulation strategies may help decrease adolescents’ aggressive communication. Moreover, female and male participants who used the digital application tended to present distinct trajectories over time with regard to searching for information concerning prosocial behavior. These findings suggest that digital tools resorting to emotion and behavioral regulation strategies may be effective in reducing an aggressive communication style amongst adolescents, and consequently, promote resource seeking to engage in prosociality. These results can be significant for the design of intervention programs against cyberbullying.

The N.C.Yeastract and CommunityYeastract databases to study gene and genomic transcription regulation in non-conventional yeasts

Abstract: Responding to the recent interest of the yeast research community in non- Saccharomyces cerevisiae species of biotechnological relevance, the N.C.Yeastract (http://yeastract-plus.org/ncyeastract/) was associated to YEASTRACT+ (http://yeastract-plus.org/). The YEASTRACT + portal is a curated repository of known regulatory associations between transcription factors (TFs) and target genes in yeasts. N.C.Yeastract gathers all published regulatory associations and transcription factor-binding sites for Komagataella phaffii (formerly Pichia pastoris), the oleaginous yeast Yarrowia lipolytica, the lactose fermenting species Kluyveromyces lactis and Kluyveromyces marxianus, and the remarkably weak acid-tolerant food spoilage yeast Zygosaccharomyces bailii. The objective of this review paper is to advertise the update of the existing information since the release of N.C.Yeastract in 2019, and to raise awareness in the community about its potential to help the day-to-day work on these species, exploring all the information available in the global YEASTRACT + portal. Using simple and widely used examples, a guided exploitation is offered for several tools: i) inference of orthologous genes; ii) search for putative transcription factor binding sites; iii) inter-species comparison of transcription regulatory networks and prediction of transcription factor-regulated networks based on documented regulatory associations available in YEASTRACT + for well-studied species. The usage potentialities of the new CommunityYeastract platform by the yeast community are also discussed.

A DC-DC Converter for Bipolar DC Microgrids with Voltage Balance Capability to Supply a Multilevel SRM Drive

Abstract: In this work it is proposed a DC-DC converter to supply a multilevel Switched Reluctance Machine (SRM) drive. The proposed converter was designed to be integrated in a bipolar DC microgrid. Usually, this kind of microgrid requires a support to avoid unbalances between the poles. Thus, the proposed converter also integrates voltage balance capabilities allowing the required support to the bipolar DC microgrid. Besides that, the converter can also supply the drive using only one of the poles. The behavior and operating modes of the converter will be described in detail. Several simulation tests will also be provided. From these tests, the behavior and the several operation modes of the DC-DC as well as the multilevel SRM drive supplied by this converter will also be presented and analyzed.

A Wideband Channel Model for Body Area Networks in Circular Metallic Indoor Environments

Abstract: In this paper, the wideband characterization of the propagation channel in circular metallic indoor environments is addressed, regarding Body Area Networks and 5G small cells, an analytical model for the dependence of the mean delay and the average delay spread on the circle radius, the working frequency and the distance between the transmitter and the receiver being proposed. The derivation of the model is initially done analytically, based on optical physics, after which simulation results allow to obtain the values of the coefficients. The simulator was previously assessed with measurements at 2.45 GHz in a passenger ferry room with a circular shape. For a random positioning of the transmitter and the receiver, and a given distance between them, it is observed that the mean delay and the delay spread increase linearly with the radius; furthermore, the mean delay increases quadratically with the distance, while the delay spread has a concave parabolic behavior, with the maximum being at a distance equal to the radius. In a practical case, regarding the positioning of an Access Point inside the room, it is recommended that it is done at the circle center, in order to reduce delay spread.

Electrification for the elite? Examining whether electrification programs reach the poor in Bolivia

Abstract: The Government of Bolivia (GoB) aims to provide electricity coverage for all citizens by 2025. This implies reaching remote areas, where electricity access rates are currently below 90% (as of 2018). To achieve this objective, the country has implemented several initiatives, such as the “Living with Dignity” program (PEVD) launched in 2008 to promote access to electricity for all, targeting the country’s poorest areas. Identifying these communities requires relying on different poverty indicators, including the unsatisfied basic needs (UBN) index. This paper focuses on identifying whether the PEVD is targeting the poorest communities. Thus, we explore the criteria that guide the GoB in selecting the areas that should be prioritised. We analyse the suitability of using the UBN index as the social criterion for this purpose. We draw on data provided by the local Censuses of 2001 and 2012. Results suggest that Bolivia’s poorest states, namely those having the highest UBN indices for 2001, experienced the largest improvements in electrification between 2001 and 2012. These results could enable policymakers to target future interventions and prioritise the poor for the provision of electricity access. However, using complementary socio-economic indicators, such as the Multidimensional poverty index (MPI), could complement the traditional UBN index by capturing more relevant information that could lead to a more accurate poverty measurement and rural electrification interventions. In addition, Bolivia is not on track to achieve universal access by 2025 and additional investments should focus on increasing electricity access rates in the states of Potosi, Pando, and Beni.

Towards better subtitles: A multilingual approach for punctuation restoration of speech transcripts

Abstract: This paper proposes a flexible approach for punctuation prediction that can be used to produce state-of-the-art results in a multilingual scenario. We have performed experiments using transcripts of TED Talks from the IWSLT 2017 and IWSLT 2011 evaluation campaigns. Our experiments show that the recognition errors of the ASR output degrade the performance of our models, in line with related literature. Our monolingual models perform consistently in Human-edited transcripts of German, Dutch, Portuguese and Romanian, suggesting that commas may be more difficult to predict than periods, using pre-trained contextual models. We have trained a single multilingual model that predicts punctuation in multiple languages that achieves results comparable with the ones achieved by monolingual models, revealing evidence of the potential of using a single multilingual model to solve the task for multiple languages. Then, we argue that usage of current punctuation systems in the literature are implicitly dependent on correct segmentation of ASR outputs for they rely on positional information to solve the punctuation task. This is too big of a requirement for use in a real life application. Through several experiments, we show that our method to train and test models is more robust to different segmentation. These contributions are of particular importance in our multilingual pipeline, since they avoid training a different model for each of the involved languages, and they guarantee that the model will be more robust to incorrect segmentation of the ASR outputs in comparison with other methods in the literature. To the best of our knowledge, we report the first experiments using a single multilingual model for punctuation restoration in multiple languages.

Assessing kinetic meaning of music and dance via deep cross-modal retrieval

Abstract: Music semantics is embodied, in the sense that meaning is biologically mediated by and grounded in the human body and brain. This embodied cognition perspective also explains why music structures modulate kinetic and somatosensory perception. We explore this aspect of cognition, by considering dance as an overt expression of semantic aspects of music related to motor intention, in an artificial deep recurrent neural network that learns correlations between music audio and dance video. We claim that, just like human semantic cognition is based on multimodal statistical structures, joint statistical modeling of music and dance artifacts is expected to capture semantics of these modalities. We evaluate the ability of this model to effectively capture underlying semantics in a cross-modal retrieval task, including dance styles in an unsupervised fashion. Quantitative results, validated with statistical significance testing, strengthen the body of evidence for embodied cognition in music and demonstrate the model can recommend music audio for dance video queries and vice versa.

From a genome assembly to full regulatory network prediction: the case study of Rhodotorula toruloides putative Haa1-regulon.

Abstract: Numerous genomes are sequenced and made available to the community through the NCBI portal. However, and, unlike what happens for gene function annotation, annotation of promoter sequences and the underlying prediction of regulatory associations is mostly unavailable, severely limiting the ability to interpret genome sequences in a functional genomics perspective. Here we present an approach where one can download a genome of interest from NCBI in the GenBank Flat File (.gbff) format and, with a minimum set of commands, have all the information parsed, organized and made available through the platform web interface. Also, the new genomes are compared with a given genome of reference in search of homologous genes, shared regulatory elements and predicted transcription associations. We present this approach within the context of Community YEASTRACT of the YEASTRACT + portal, thus benefiting from immediate access to all the comparative genomics queries offered in the YEASTRACT + portal. Besides the yeast community, other communities can install the platform independently, without any constraints. In this work, we exemplify the usefulness of the presented tool, within Community YEASTRACT, in constructing a dedicated database and analysing the genome of the highly promising oleaginous red yeast species Rhodotorula toruloides currently poorly studied at the genome and transcriptome levels and with limited genome editing tools. Regulatory prediction is based on the conservation of promoter sequences and available regulatory networks. The case-study examined is focused on the Haa1 transcription factor-a key regulator of yeast resistance to acetic acid, an important inhibitor of industrial bioconversion of lignocellulosic hydrolysates. The new tool described here led to the prediction of a RtHaa1 regulon with expected impact in the optimization of R. toruloides robustness for lignocellulosic and pectin-rich residue biorefinery processes.

Fast and energy-efficient approximate motion estimation architecture for real-time 4 K UHD processing

Abstract: Approximate computing techniques exploit the characteristics of error-tolerant applications either to provide faster implementations of their computational structures or to achieve substantial improvements in terms of energy efficiency. In video encoding, the motion estimation (ME) stage, including the Integer ME (IME) and the Fractional ME (FME) steps, is the most computational intensive task and it is highly resilient to controlled losses of accuracy. In accordance, this article proposes the exploitation of approximate computing techniques to implement energy efficient dedicated hardware structures targeting the motion estimation stage of current video encoders. The designed ME architecture supports IME and FME and is able to real-time process 4 K UHD videos (3840 × 2160 pixels) at 30 frames per second, while dissipating 108.92 mW. When running at its maximum operation frequency, the architecture can process 8 K UHD videos (7680 × 4320 pixels) at 120 frames per second. The solution described in this article presents the highest throughput and the highest energy efficiency among all state-of-the-art compared works, showing that the use of approximate computing is a promising solution when implementing video encoders in dedicated hardware.

Quantum-like Gaussian mixture model

Abstract: A new concept of a quantum-like mixture model is introduced. It describes the mixture distribution with the assumption that a point is generated by each Gaussian at the same time. The quantum-like mixture Gaussian improves the classification accuracy in machine learning by indicating that the uncertain points should not be assigned to any class. It increases the accuracy of the mixture Gaussian model on the iris data set from 96.67 to 99.24%.

Using brain inspired principles to unsupervisedly learn good representations for visual pattern recognition.

Abstract: Although deep learning has solved difficult problems in visual pattern recognition, it is mostly successful in tasks where there are lots of labeled training data available. Furthermore, the global back-propagation based training rule and the amount of employed layers represents a departure from biological inspiration. The brain is able to perform most of these tasks in a very general way from limited to no labeled data. For these reasons it is still a key research question to look into computational principles in the brain that can help guide models to unsupervisedly learn good representations which can then be used to perform tasks like classification. In this work we explore some of these principles to generate such representations for the MNIST data set. We compare the obtained results with similar recent works and verify extremely competitive results.

Development of an off-grid micro-inverter for a scalable photovoltaic kit

Abstract: Taking into account the geographic location of some rural areas, it is not viable, due to technical or economical reasons, to connect certain villages to the electrical grid. Since access to energy is essential for human development, a DC-DC-AC micro-inverter topology for a photovoltaic kit able to address the previous scenario is proposed in this paper. This kit is capable of working in both isolated and parallel modes, with other kits. This solution is also adequate for a post-natural catastrophe scenario, given the power outages associated with these types of phenomena. Each kit consists of a photovoltaic module, an energy storage system and a DC-DC-AC micro-inverter. A control system for all the subsystems is also presented in this paper. Besides that, the management of the DC bus, taking into consideration the photovoltaic system and energy storage system is also proposed. All the theoretical analysis of this kit is detailed in the paper and confirmed through simulations.

Spectrum-based feature localization: a case study using ArgoUML

Abstract: Feature localization (FL) is a basic activity in re-engineering legacy systems into software product lines. In this work, we explore the use of the Spectrum-based localization technique for this task. This technique is traditionally used for fault localization but with practical applications in other tasks like the dynamic FL approach that we propose. The ArgoUML SPL benchmark is used as a case study and we compare it with a previous hybrid (static and dynamic) approach from which we reuse the manual and testing execution traces of the features. We conclude that it is feasible and sound to use the Spectrum-based approach providing promising results in the benchmark metrics.

On the predictability of postoperative complications for cancer patients: a Portuguese cohort study

Abstract: Postoperative complications are still hard to predict despite the efforts towards the creation of clinical risk scores. The published scores contribute for the creation of specialized tools, but with limited predictive performance and reusability for implementation in the oncological context. This work aims to predict postoperative complications risk for cancer patients, offering two major contributions. First, to develop and evaluate a machine learning-based risk score, specific for the Portuguese population using a retrospective cohort of 847 cancer patients undergoing surgery between 2016 and 2018, for 4 outcomes of interest: (1) existence of postoperative complications, (2) severity level of complications, (3) number of days in the Intermediate Care Unit (ICU), and (4) postoperative mortality within 1 year. An additional cohort of 137 cancer patients from the same center was used for validation. Second, to improve the interpretability of the predictive models. In order to achieve these objectives, we propose an approach for the learning of risk predictors, offering new perspectives and insights into the clinical decision process. For postoperative complications the Receiver Operating Characteristic Curve (AUC) was 0.69, for complications’ severity AUC was 0.65, for the days in the ICU the mean absolute error was 1.07 days, and for 1-year postoperative mortality the AUC was 0.74, calculated on the development cohort. In this study, predictive models which could help to guide physicians at organizational and clinical decision making were developed. Additionally, a web-based decision support tool is further provided to this end.

Helping People on the Fly: Ad Hoc Teamwork for Human-Robot Teams.

Abstract: We present the Bayesian Online Prediction for Ad hoc teamwork (BOPA), a novel algorithm for ad hoc teamwork which enables a robot to collaborate, on the fly, with human teammates without any pre-coordination protocol. Unlike previous works, BOPA relies only on state observations/transitions of the environment in order to identify the task being performed by a given teammate (without observing the teammate’s actions and environment’s reward signals). We evaluate BOPA in two distinct settings, namely (i) an empirical evaluation in a simulated environment with three different types of teammates, and (ii) an experimental evaluation in a real-world environment, deploying BOPA into an ad hoc robot with the goal of assisting a human teammate in completing a given task. Our results show that BOPA is effective at correctly identifying the target task, efficient at solving the correct task in optimal and near-optimal times, scalable by adapting to different problem sizes, and robust to non-optimal teammates, such as humans.

DI2: prior-free and multi-item discretization of biological data and its applications.

Abstract: BACKGROUND A considerable number of data mining approaches for biomedical data analysis, including state-of-the-art associative models, require a form of data discretization. Although diverse discretization approaches have been proposed, they generally work under a strict set of statistical assumptions which are arguably insufficient to handle the diversity and heterogeneity of clinical and molecular variables within a given dataset. In addition, although an increasing number of symbolic approaches in bioinformatics are able to assign multiple items to values occurring near discretization boundaries for superior robustness, there are no reference principles on how to perform multi-item discretizations. RESULTS In this study, an unsupervised discretization method, DI2, for variables with arbitrarily skewed distributions is proposed. Statistical tests applied to assess differences in performance confirm that DI2 generally outperforms well-established discretizations methods with statistical significance. Within classification tasks, DI2 displays either competitive or superior levels of predictive accuracy, particularly delineate for classifiers able to accommodate border values. CONCLUSIONS This work proposes a new unsupervised method for data discretization, DI2, that takes into account the underlying data regularities, the presence of outlier values disrupting expected regularities, as well as the relevance of border values. DI2 is available at https://github.com/JupitersMight/DI2.