Large Task Sets Research Articles

Preparing of Youngest Students for Participation in Programming Contests

The goal of this paper is to present in brief the analysis of large set of tasks from Bulgarian national and regional programming contests for age group E (4th–5th grades), published in (Manev, 2023). As a result of the analysis, the identified in more than 350 tasks topics are arranged in order of smoothly increasing difficulty and could be used as training curriculum for preparation of programming contestants of early age.

BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition

We summarize the results of a host of efforts using giant automatic speech recognition (ASR) models pre-trained using large, diverse unlabeled datasets containing approximately a million hours of audio. We find that the combination of pre-training, self-training and scaling up model size greatly increases data efficiency, even for extremely large tasks with tens of thousands of hours of labeled data. In particular, on an ASR task with 34k hours of labeled data, by fine-tuning an 8 billion parameter pre-trained Conformer model we can match state-of-the-art (SoTA) performance with only 3% of the training data and significantly improve SoTA with the full training set. We also report on the universal benefits gained from using big pre-trained and self-trained models for a large set of downstream tasks that cover a wide range of speech domains and span multiple orders of magnitudes of dataset sizes, including obtaining SoTA performance on many public benchmarks. In addition, we utilize the learned representation of pre-trained networks to achieve SoTA results on non-ASR tasks.

Automated Open-Hardware Multiwell Imaging Station for Microorganisms Observation.

Bright field microscopes are particularly useful tools for biologists for cell and tissue observation, phenotyping, cell counting, and so on. Direct cell observation provides a wealth of information on cells’ nature and physiological condition. Microscopic analyses are, however, time-consuming and usually not easy to parallelize. We describe the fabrication of a stand-alone microscope able to automatically collect samples with 3D printed pumps, and capture images at up to 50× optical magnification with a digital camera at a good throughput (up to 24 different samples can be collected and scanned in less than 10 min). Furthermore, the proposed device can store and analyze pictures using computer vision algorithms running on a low power integrated single board computer. Our device can perform a large set of tasks, with minimal human intervention, that no single commercially available machine can perform. The proposed open-hardware device has a modular design and can be freely reproduced at a very competitive price with the use of widely documented and user-friendly components such as Arduino, Raspberry pi, and 3D printers.

Systematic search space design for energy-efficient static scheduling of moldable tasks

Static scheduling of independent, moldable tasks on parallel machines with frequency scaling comprises decisions on core allocation, assignment, frequency scaling and ordering, to meet a deadline and minimize energy consumption. Constraining some of these decisions reduces the solution space, i.e. may increase energy consumption, but may also open the path to new, near-optimal approaches. We investigate how constraints of different steps influence energy consumption, starting with an unrestricted scheduler for moldable tasks. The constraints are partly derived from existing schedulers, but also generalized in a systematic way. We present integer linear programs for all scheduling variants. We compare energy consumption of schedules for a benchmark suite of synthetic task sets of different sizes and for task sets derived from real applications. In addition, we check how close the results are to the optimum results when the ILP solver meets a timeout. Our results indicate that constraints on task execution order, which avoid explicit representation of task order in ILPs, are mostly responsible for near-optimal energy consumption among large task sets. Furthermore, we find that for all steps except allocation, non-optimal fast heuristics can be used without sacrificing too much energy for the resulting schedule. Finally, we can show that an ILP for a new scheduler, for which also a heuristic version exists, is comparable in quality to more complicated schedulers.

A Framework for Indoor Positioning Including Building Topology

In many application domains, position information is of fundamental importance. However, unlike the case of outdoor positioning, producing an accurate position estimation in the indoor setting turns out to be quite difficult. One of the most common localisation strategies makes use of fingerprinting. Research in this area has been faced with a number of challenges, leading to the proposal of a number of localisation algorithms, sampling strategies, benchmark datasets, and representations of building information. This proliferation made the modeling of the indoor positioning domain quite hard from both a theoretical and a practical point of view. In this paper, we propose a general and extensible framework, based on a relational database, that pairs fingerprints with building information.We show how the proposed system successfully deals with a number of problems that affect indoor positioning, supporting a large set of relevant tasks. The source code of the framework is available online, as well as an implementation of it, that provides an interactive open repository of indoor positioning data.

Evaluation of Task Scheduling Algorithms in Heterogeneous Computing Environments.

This work establishes a set of methodologies to evaluate the performance of any task scheduling policy in heterogeneous computing contexts. We formally state a scheduling model for hybrid edge–cloud computing ecosystems and conduct simulation-based experiments on large workloads. In addition to the conventional cloud datacenters, we consider edge datacenters comprising smartphone and Raspberry Pi edge devices, which are battery powered. We define realistic capacities of the computational resources. Once a schedule is found, the various task demands can or cannot be fulfilled by the resource capacities. We build a scheduling and evaluation framework and measure typical scheduling metrics such as mean waiting time, mean turnaround time, makespan, throughput on the Round-Robin, Shortest Job First, Min-Min and Max-Min scheduling schemes. Our analysis and results show that the state-of-the-art independent task scheduling algorithms suffer from performance degradation in terms of significant task failures and nonoptimal resource utilization of datacenters in heterogeneous edge–cloud mediums in comparison to cloud-only mediums. In particular, for large sets of tasks, due to low battery or limited memory, more than 25% of tasks fail to execute for each scheduling scheme.

Recommending Tasks in Online Judges using Autoencoder Neural Networks

Programming contests such as International Olympiads in Informatics (IOI) and ACM International Collegiate Programming Contest (ICPC) are becoming increasingly popular in recent years. To train for these contests, there are several Online Judges available, in which users can test their skills against a usually large set of programming tasks. In the literature, so far few papers have addressed the problem of recommending tasks in online judges. Most notably, as opposed with traditional Recommender Systems, since the learners improve their skills as they solve more problems, there is an intrinsic dynamic dimension that has to be considered: when recommending movies or books, it is likely that the preferences of the users are more or less stable, whilst in recommending tasks this does not hold true. In order to help the learners, it is crucial to recommend them tasks that are challenging but not unsolvable compared with their current set of skills. In this paper we present a Recommender System (RS) for Online Judges based on an Autoencoder (Artificial) Neural Network (ANN). We also discuss the results of an experimental evaluation of our approach in both the scenarios in which we consider, or not, the intrinsic dynamic dimension of the problem. The ANNs are trained with the dataset of all the submissions in the Italian National Online Judge, used to train students for the Italian Olympiads in Informatics.

A Comprehensive Spatial Mapping of Muscle Synergies in Highly Variable Upper-Limb Movements of Healthy Subjects.

BackgroundRecently, muscle synergy analysis has become a standard methodology for extracting coordination patterns from electromyographic (EMG) signals, and for the evaluation of motor control strategies in many contexts. Most previous studies have characterized upper-limb muscle synergies across a limited set of reaching movements. With the aim of future uses in motor control, rehabilitation and other fields, this study provides a comprehensive characterization of muscle synergies in a large set of upper-limb tasks and also considers inter-individual and environmental variability.MethodsSixteen healthy subjects performed upper-limb hand exploration movements for a comprehensive mapping of the upper-limb workspace, which was divided into several sectors (Frontal, Right, Left, Horizontal, and Up). EMGs from representative upper-limb muscles and kinematics were recorded to extract muscle synergies and explore the composition, repeatability and similarity of spatial synergies across subjects and movement directions, in a context of high variability of motion.ResultsEven in a context of high variability, a reduced set of muscle synergies may reconstruct the original EMG envelopes. Composition, repeatability and similarity of synergies were found to be shared across subjects and sectors, even if at a lower extent than previously reported.ConclusionExtending the results of previous studies, which were performed on a smaller set of conditions, a limited number of muscle synergies underlie the execution of a large variety of upper-limb tasks. However, the considered spatial domain and the variability seem to influence the number and composition of muscle synergies. Such detailed characterization of the modular organization of the muscle patterns for upper-limb control in a large variety of tasks may provide a useful reference for studies on motor control, rehabilitation, industrial applications, and sports.

Understanding inclusion in teacher education – a view from student teachers in England

ABSTRACT Policies on inclusion are being increasingly embedded within education systems and teacher education across the world, with schools and teachers called upon to add ‘inclusion’ to their already large set of skills and tasks. There is, however, no consistent definition of what inclusion means or how it can be best promoted. The purpose of this paper is to explore the dilemmas that student teachers face when they encounter policy requirements to practice inclusion, and how they mediate the tensions. Drawing on two exploratory studies with science student teachers in two Initial Teacher Education programmes in England, we focus on the conceptions of inclusion held by the student teachers and the links between inclusion and teacher education. Our findings suggest that conventional understandings in relation to ability still dominate, with ability-based differentiation viewed as the key teaching strategy to promote inclusion. In addition, student teachers find themselves having to negotiate contradictory and often conflicting approaches to inclusion, diversity, and academic attainment. The discrepancies highlighted by this study have implications for how teacher education courses need to be organised to promote the practice of inclusion.

Incremental Learning Introspective Movement Primitives From Multimodal Unstructured Demonstrations

Learning movement primitive from unstructured demonstrations has become a popular topic in recent years, which provides a natural way to endow human-inspired skills to robots. The main idea of movement primitives is that should suffice to reconstruct a large set of complex manipulation tasks. However, conventional learning methods mostly focus on the kinesthetic variables and ignore those critical introspective capacities in manipulation such as movement generalization and assessment of the sensory signals. In this paper, we investigate the association of generalization, fault detection, fault diagnoses, and task exploration during manipulation task, and call such movement primitives augmented with introspective capacities Introspective Movement Primitives (IMP). With our previous work, this paper mainly addresses how IMPs can be acquired by assessing the quality of multimodal sensory data of unstructured demonstrations and how they can incrementally create manipulation task by reverse execution and human interaction. Experimental evaluation on a human-robot collaborative packaging task with a Rethink Baxter robot, results indicate that our proposed method can effectively increase robustness towards external perturbations and adaptive exploration during robot manipulation task.

Schedulability Analysis of Graph-Based Real-Time Task Model with Precedence Constraints

The timing requirements of real-time systems can be guaranteed by well-designed scheduling policies. The analysis of such scheduling uses an abstract task model of the system to diagnose the practicability of timing requirements. The task models have evolved from periodic models to more sophisticated graph-based ones, among which digraph real-time (DRT) task model is the most applicable because of its good expressiveness and analysis efficiency. However, the DRT model cannot support the commonly used precedence constraints within or between tasks. In this paper, we propose a new task model that extends the DRT model to support precedence constraints. Based on our model, we present two methods of uniprocessor schedulability analysis for static priority scheduling policy and earliest deadline first (EDF) scheduling policy. We also introduce an optimization technique to improve the efficiency of model analysis. Our experiments show that, despite a high computational complexity of the problem, our approach scales very well for large sets of tasks with precedence constraints.

Task Assignment Model for Crowdsourcing Software Development: TAM

Selection of a suitable task from the extensively available large set of tasks is an intricate job for the developers in crowdsourcing software development (CSD). Besides, it is also a tiring and a time-consuming job for the platform to evaluate thousands of tasks submitted by developers. Previous studies stated that managerial and technical aspects have prime importance in bringing success for software development projects, however, these two aspects can be more effective and conducive if combined with human aspects. The main purpose of this paper is to present a conceptual framework for task assignment model for future research on the basis of personality types, that will provide a basic structure for CSD workers to find suitable tasks and also a platform to assign the task directly. This will also match their personality and task. Because personality is an internal force which whittles the behavior of developers. Consequently, this research presented a Task Assignment Model (TAM) from a developers point of view, moreover, it will also provide an opportunity to the platform to assign a task to CSD workers according to their personality types directly.

A Unifying View on SMT-Based Software Verification

After many years of successful development of new approaches for software verification, there is a need to consolidate the knowledge about the different abstract domains and algorithms. The goal of this paper is to provide a compact and accessible presentation of four SMT-based verification approaches in order to study them in theory and in practice. We present and compare the following different “schools of thought” of software verification: bounded model checking, k-induction, predicate abstraction, and lazy abstraction with interpolants. Those approaches are well-known and successful in software verification and have in common that they are based on SMT solving as the back-end technology. We reformulate all four approaches in the unifying theoretical framework of configurable program analysis and implement them in the verification framework CPAchecker. Based on this, we can present an evaluation that thoroughly compares the different approaches, where the core differences are expressed in configuration parameters and all other variables are kept constant (such as parser front end, SMT solver, used theory in SMT formulas). We evaluate the effectiveness and the efficiency of the approaches on a large set of verification tasks and discuss the conclusions.

Trajectory Generation for Unmanned Aerial Manipulators Through Quadratic Programming

In this paper, a trajectory generation approach using quadratic programming is described for aerial manipulation, i.e. , for the control of an aerial vehicle equipped with a robot arm. The proposed approach applies the online active set strategy to generate a feasible trajectory of the joints, in order to accomplish a set of tasks with defined bounds and constraint inequalities. The definition of the problem in the acceleration domain allows to integrate and perform a large set of tasks and, as a result, to obtain smooth motion of the joints. A weighting strategy, associated with a normalization procedure, allows us to easily define the relative importance of the tasks. This approach is useful to accomplish different phases of a mission with different redundancy resolution strategies. The performance of the proposed technique is demonstrated through real experiments with all the algorithms running onboard in real time. In particular, the aerial manipulator can successfully perform navigation and interaction phases, while keeping motion within prescribed bounds and avoiding collisions with external obstacles.

A Light-Weight Approach for Verifying Multi-Threaded Programs with CPAchecker

Verifying multi-threaded programs is becoming more and more important, because of the strong trend to increase the number of processing units per CPU socket. We introduce a new configurable program analysis for verifying multi-threaded programs with a bounded number of threads. We present a simple and yet efficient implementation as component of the existing program-verification framework CPAchecker. While CPAchecker is already competitive on a large benchmark set of sequential verification tasks, our extension enhances the overall applicability of the framework. Our implementation of handling multiple threads is orthogonal to the abstract domain of the data-flow analysis, and thus, can be combined with several existing analyses in CPAchecker, like value analysis, interval analysis, and BDD analysis. The new analysis is modular and can be used, for example, to verify reachability properties as well as to detect deadlocks in the program. This paper includes an evaluation of the benefit of some optimization steps (e.g., changing the iteration order of the reachability algorithm or applying partial-order reduction) as well as the comparison with other state-of-the-art tools for verifying multi-threaded programs.

Control-Scheduling Codesign Exploiting Trade-Off between Task Periods and Deadlines

A control task’s performance heavily depends on its sampling frequency and sensing-to-actuation delay. More frequent sampling, that is, shorter period, improves the control performance. Similarly, shorter delay also has a positive effect. Moreover, schedulability is also a function of periods and deadlines. By taking into account the control performance and schedulability at the same time, this paper defines a period and deadline selection problem for fixed-priority systems. Our problem is to find the optimal periods and deadlines for given tasks that maximize the overall system performance. As our solution, this paper presents a novel heuristic algorithm that finds a high-quality suboptimal solution with very low complexity, which makes the algorithm practically applicable to large size task sets.

Is it harder to switch among a larger set of tasks?

When stimuli afford multiple tasks, switching among them involves promoting one of several task-sets in play into a most-active state. This process, often conceptualized as retrieving task parameters and stimulus-response (S-R) rules into procedural working memory, is a likely source of the reaction time (RT) cost of a task-switch, especially when no time is available for task preparation before the stimulus. We report 2 task-cuing experiments that asked whether the time consumed by task-set retrieval increases with the number of task-sets in play, while unconfounding the number of tasks with their frequency and recency of use. Participants were required to switch among 3 or 5 orthogonal classifications of perceptual attributes of an object (Experiment 1) or of phonological/semantic attributes of a word (Experiment 2), with a 100 or 1,300 ms cue-stimulus interval. For 2 tasks for which recency and frequency were matched in the 3- and 5-task conditions, there was no effect of number of tasks on the switch cost. For the other tasks, there was a greater switch cost in the 5-task condition with little time for preparation, attributable to effects of frequency/recency. Thus, retrieval time for active task-sets is not influenced by the number of alternatives per se (unlike several other kinds of memory retrieval) but is influenced by recency or frequency of use.

Data and indicators for evidence-based livestock sector policies and investments.

Abstract Livestock sector policies' and investments' effectiveness is enhanced by their evidence base. However, the lack of mainstream literature on the information needs of decision-makers makes it difficult to appreciate the data and indicators necessary for effective policies and investments. This review systematizes the types of data and indicators that decision-makers need for effectively formulating and implementing policies and investments in the livestock sector through the various phases of the decision-making process. These span the large set of tasks associated with assessing the contribution of livestock to development, to identifying priority areas of investments, to monitoring and evaluation. Six domains for such data are identified and discussed, with key contributions from the literature identified. Basic data and indicators provide critical information to decision-makers, with statistical analyses linking inputs and outputs further improving the decision-making processes. However, these are rarely readily available to draft an evidence-based policy implementation plan. Indeed, the implementation of policy reforms and investments usually co-occurs with some form of institutional change - new ways of doing things - which calls for changed behaviours for both implementers and beneficiaries. Elements of a statistical system that is largely effective in promoting sustainable development of the livestock sector generally is one that generates good quality livestock data and indicators representative of the country as a whole and major livestock production systems, uses inclusive participatory policy processes and consultation with experts, synthesizes existing experience and analysis, and uses both rigorous ex ante analysis and pilots to assist decision-makers in designing and executing policies and investments, and implementing monitoring and evaluation.

An Optimized Algorithm for Enhancement of Performance of Distributed Computing System

Distributed Computing System (DCS) presents a platform consisting of multiple computing nodes connected in some fashion to which various modules of a task can be assigned. A node is any device connected to a computer network. Nodes can be computers or various other network applications. A task should be assigned to a processor whose capabilities are most appropriate for the execution of that task. In a DCS, a number of tasks are allocated to different processors in such a way that the overall performance in terms of time, cost should be minimized and reliability should be maximized. For a large set of tasks that is being allocated into a DCS, several allocation methods are possible. These allocations can have significant impact on quality of services such as time, cost or reliability. Execution time is the time in which a single instruction is executed. Execution cost can be termed as the amount of value of resource used. In DCS reliability is highly dependent on its network and failures of network have adverse impact on the system performance. In DCS the whole workload is divided into small and independent units, called tasks and it allocates onto the available processors. In this paper a simple algorithm for task allocation in terms of optimum time or optimum cost or optimum reliability is presented where the numbers of tasks are more then the number of processors.

Effects of Usage-Based Feedback on Video Retrieval

We present a model for exploiting community-based usage information for video retrieval, where implicit usage information from past users is exploited in order to provide enhanced assistance in video retrieval tasks, and alleviate the effects of the semantic gap problem. We propose a graph-based model for all types of implicit and explicit feedback, in which the relevant usage information is represented. Our model is designed to capture the complex interactions of a user with an interactive video retrieval system, including the representation of sequences of user-system interaction during a search session. Building upon this model, four recommendation strategies are defined and evaluated. An evaluation strategy is proposed based on simulated user actions, which enables the evaluation of our recommendation strategies over a usage information pool obtained from 24 users performing four different TRECVid tasks. Furthermore, the proposed simulation approach is used to simulate usage information pools with different characteristics, with which the recommendation approaches are further evaluated on a larger set of tasks, and their performance is studied with respect to the scalability and quality of the available implicit information.