Active Integrity Constraints for Database Consistency Maintenance

This paper introduces active integrity constraints (AICs), an extension of integrity constraints for consistent database maintenance. An active integrity constraint is a special constraint whose body contains a conjunction of literals which must be false and whose head contains a disjunction of update actions representing actions (insertions and deletions of tuples) to be performed if the constraint is not satisfied (that is its body is true). The AICs work in a domino-like manner as the satisfaction of one AIC may trigger the violation and therefore the activation of another one. The paper also introduces founded repairs, which are minimal sets of update actions that make the database consistent, and are specified and “supported” by active integrity constraints. The paper presents: 1) a formal declarative semantics allowing the computation of founded repairs and 2) a characterization of this semantics obtained by rewriting active integrity constraints into disjunctive logic rules, so that founded repairs can be derived from the answer sets of the derived logic program. Finally, the paper studies the computational complexity of computing founded repairs.     

The influence of modeling hypothesis and experimental methodologies in the accuracy of muscle force estimation using EMG-driven models

This paper discusses some methodological questions regarding the application of EMG-driven models to estimate muscle forces, for the triceps surae performing isometric contractions. Ankle torque is estimated from a Hill-type muscle model driven by EMG data, collected from the three components of triceps surae and tibialis anterior. Ankle joint torque is synchronously collected from a dynamometer, which is compared to the sum of each muscle force multiplied by the respective ankle moment arm. A protocol consisting of two steps of low and medium/high loads is used. Raw EMG signal is processed and used as the input signal for the muscle model. The difference between simulated and dynamometer measured torque is calculated as the RMS error between the two curves. A set of nominal muscle model parameters is initially chosen from literature (e.g., OpenSim), which allows observing the characteristics of the error distribution. One possibility to improve model accuracy is using individual muscle parameters. We investigated the effect of applying simple scale factors to the nominal muscle model parameters and using ultrasound for estimating muscle maximum force. Other questions regarding muscle model improvements are also addressed, such as using a nonlinear formulation of activation dynamics and variable pennation angle. Surface EMG signals acquisition and processing can also affect force estimation accuracy. Electrodes positioning can influence signal amplitude, and the one-channel EMG may not represent actual excitation for the whole muscle. We have shown that high density EMG reduces, in some cases, the torque estimation error.     

Sharing beliefs among agents with different degrees of credibility

This paper introduces an approach for sharing beliefs in collaborative multi-agent application domains where some agents can be more credible than others. In this context, we propose a formalization where every agent has its own partial order among its peers representing the credibility the agent assigns to its informants; each agent will also have a belief base where each sentence is attached with an agent identifier which represents the credibility of that sentence. We define four different forwarding criteria for computing the credibility information for a belief to be forwarded, and for determining how the receiver should handle the incoming information; the proposal considers both the sender’s and the receiver’s points of view with respect to the credibility of the source of the information.     

On computing the entropy of cellular automata

We study the topological entropy of a particular class of dynamical systems: cellular automata. The topological entropy of a dynamical system (X,F) is a measure of the complexity of the dynamics of F over the space X. The problem of computing (or even approximating) the topological entropy of a given cellular automata is algorithmically undecidable (Ergodic Theory Dynamical Systems 12 (1992) 255). In this paper, we show how to compute the entropy of two important classes of cellular automata namely, linear and positively expansive cellular automata. In particular, we prove a closed formula for the topological entropy of D-dimensional (D1) linear cellular automata over the ring Z_m (m2) and we provide an algorithm for computing the topological entropy of positively expansive cellular automata.     

Ultimate strength analysis of prestressed reinforced concrete sections under axial force and biaxial bending

A secant approach is illustrated for the ultimate limit state (ULS) analysis of prestressed reinforced concrete sections subjected to axial load and biaxial bending in presence of softening stress–strain laws. The stiffness matrix and the resultant loads are evaluated analytically by a novel methodology, termed fiber-free, which represents a computationally efficient alternative to fiber approaches. Extensive computations of the ULS domains of benchmark test cases show that the robustness of the proposed algorithmic strategy is substantially unaffected by the amount of reinforcement, prestressing and softening, though localized non-convex regions have been occasionally experienced in presence of softening.     

Bootstrap analysis of multiple repetitions of experiments using an interval-valued multiple comparison procedure

A new bootstrap test is introduced that allows for assessing the significance of the differences between stochastic algorithms in a cross-validation with repeated folds experimental setup. Intervals are used for modeling the variability of the data that can be attributed to the repetition of learning and testing stages over the same folds in cross validation. Numerical experiments are provided that support the following three claims: (1) Bootstrap tests can be more powerful than ANOVA or Friedman test for comparing multiple classifiers. (2) In the presence of outliers, interval-valued bootstrap tests achieve a better discrimination between stochastic algorithms than nonparametric tests. (3) Choosing ANOVA, Friedman or Bootstrap can produce different conclusions in experiments involving actual data from machine learning tasks.     

A reduced memory bandwidth and high throughput HDTV motion compensation decoder for H.264/AVC High 4:2:2 profile

This article presents the HP422-MoCHA: optimized Motion Compensation hardware architecture for the High 4:2:2 profile of H.264/AVC video coding standard. The proposed design focuses on real-time decoding for HDTV 1080p (1,920 × 1,080 pixels) at 30 fps. It supports multiple sample bit-width (8, 9, or 10 bits) and multiple chroma sub-sampling formats (4:0:0, 4:2:0, and 4:2:2) to provide enhanced video quality experience. The architecture includes an optimized sample interpolator that processes luma and chroma samples in two parallel datapaths and features quarter sample accuracy, bi-prediction and weighted prediction. HP422-MoCHA also includes a hardwired Motion Vector Predictor, supporting temporal and spatial direct predictions. A novel memory hierarchy implemented as a 3-D Cache reduces the frame memory access, providing, on average, 62% of bandwidth and 80% of clock cycles reduction. The design was implemented in a Xilinx Virtex-II PRO FPGA, and also in an ASIC with a TSMC 0.18 μm standard cells technology. The ASIC implementation occupies 102 K equivalent gates and 56.5 KB of on-chip SRAM in a 3.8 × 3.4 mm² area. It presents a power consumption of 130 mW. Both implementations reach a maximum operation frequency of ~100 MHz, being able to motion compensate 37 bi-predictive frames or 69 predictive fps. The minimum required frequency to ensure the real-time decoding for HD1080p at 30 fps is 82 MHz. Since HP422-MoCHA is the first Motion Compensation architecture for the High 4:2:2 profile found in the literature, a Main profile MoCHA was used for comparison purposes, showing the highest throughput among all presented works. However, the HP422-MoCHA architecture also reaches the highest throughput when compared with the other published Main profile MC solutions, even considering the significantly higher complexity of the High 4:2:2 profile.     

Iterative Decentralized Planning for Collective Construction Tasks with Quadrotors

This paper describes an iterative decentralized planning and learning method, based on stochastic learning automata theory and heuristic search techniques, to generate construction and motion strategies to build different types of three-dimensional structures using multiple quadrotors. This architecture is proposed to simultaneously solve three main problems: 1) the iterative generation of feasible construction and motion plans for each quadrotor; 2) the optimization with constraints on power and assembly while taking into account the dynamic nature of the environment, and 3) the planning of the translational speeds and selection of breakpoints for each vehicle. The quadrotors learn the optimal action policy to construct the structures while avoiding collisions during the loading and unloading procedures. In order to demonstrate the generality of the solution, simulated trials of the proposed autonomous construction system are presented where different three-dimensional structures are built.     

AI4People—An Ethical Framework for a Good AI Society: Opportunities,Risks, Principles,and Recommendations

This article reports the findings of AI4People, an Atomium—EISMD initiative designed to lay the foundations for a “Good AI Society”. We introduce the core opportunities and risks of AI for society; present a synthesis of five ethical principles that should undergird its development and adoption; and offer 20 concrete recommendations—to assess, to develop, to incentivise, and to support good AI—which in some cases may be undertaken directly by national or supranational policy makers, while in others may be led by other stakeholders. If adopted, these recommendations would serve as a firm foundation for the establishment of a Good AI Society.     

Exact and Approximate Truthful Mechanisms for the Shortest Paths Tree Problem

Let a communication network be modeled by an undirected graph G=(V,E) of n nodes and m edges, and assume that edges are controlled by selfish agents. In this paper we analyze the problem of designing a truthful mechanism for computing one of the most popular structures in communication networks, i.e., the single-source shortest paths tree. More precisely, we will study several realistic scenarios, in which each agent can own either a single or multiple edges of G. In particular, for the single-edge case, we will show that: (i) in the classic utilitarian case, the problem can be solved efficiently in O(mnlog α(m,n)) time, where α(m,n) is the inverse of the Ackermann’s function; (ii) in a meaningful non-utilitarian case, namely that in which agents’ valuation functions only depend on the edge lengths, the problem can be solved in O(m+nlog n) time. Conversely, for the multiple-edges case, we will show in the utilitarian case an O(mP+nPlog n) time truthful mechanism, where P=O(n) denotes the number of agents participating in the solution, while in the same non-utilitarian case we will prove a general lower bound to the approximation ratio that can be achieved by any truthful mechanism, by showing that no c-approximate mechanism can exist, for any fixed . Work partially supported by the Research Project GRID.IT, funded by the Italian Ministry of Education, University and Research. Part of the results herein contained was presented at the 11th International Euro-Par Conference (Euro-Par’05), Lisbon, Portugal, 2005.