Search

Binary Convolutional Neural Networks (CNNs) can significantly reduce the number of arithmetic operations and the size of memory storage, which makes the deployment of CNNs on mobile or embedded systems more promising. However, the accuracy degradation of single and multiple binary CNNs is unacceptable for modern architectures and large scale datasets like ImageNet. In this paper, we proposed a Piecewise Approximation (PA) scheme for multiple binary CNNs which lessens accuracy loss by approximating full precision weights and activations efficiently, and maintains parallelism of bitwise operations to guarantee efficiency. Unlike previous approaches, the proposed PA scheme segments piece-wisely the full precision weights and activations, and approximates each piece with a scaling coefficient. Our implementation on ResNet with different depths on ImageNet can reduce both Top-1 and Top-5 classification accuracy gap compared with full precision to approximately 1.0%. Benefited from the binarization of the downsampling layer, our proposed PA-ResNet50 requires less memory usage and two times Flops than single binary CNNs with 4 weights and 5 activations bases. The PA scheme can also generalize to other architectures like DenseNet and MobileNet with similar approximation power as ResNet which is promising for other tasks using binary convolutions. The code and pretrained models will be publicly available.

While current General Game Playing (GGP) systems facilitate useful research in Artificial Intelligence (AI) for gameplaying, they are often somewhat specialised and computationally inefficient. In this paper, we describe the “ludemic” general game system Ludii, which has the potential to provide an efficient tool for AI researchers as well as game designers, historians, educators and practitioners in related fields. Ludii defines games as structures of ludemes – high-level, easily understandable game concepts – which allows for concise and human-understandable game descriptions. We formally describe Ludii and outline its main benefits: generality, extensibility, understandability and efficiency. Experimentally, Ludii outperforms one of the most efficient Game Description Language (GDL) reasoners, based on a propositional network, in all games available in the Tiltyard GGP repository. Moreover, Ludii is also competitive in terms of performance with the more recently proposed Regular Boardgames (RBG) system, and has various advantages in qualitative aspects such as generality.

The unification of low-level perception and high-level reasoning is a long-standing problem in artificial intelligence, which has the potential to not only bring the areas of logic and learning closer together but also demonstrate how abstract concepts might emerge from sensory data. Precisely because deep learning methods dominate perception-based learning, including vision, speech, and linguistic grammar, there is fast-growing literature on how to integrate symbolic reasoning and deep learning. Broadly, efforts seem to fall into three camps: those focused on defining a logic whose formulas capture deep learning, ones that integrate symbolic constraints in deep learning, and others that allow neural computations and symbolic reasoning to co-exist separately, to enjoy the strengths of both worlds. In this paper, we identify another dimension to this inquiry: what do the hidden layers really capture, and how can we reason about that logically? In particular, we consider variational autoencoders that are widely used for dimensionality reduction and inject a symbolic generative framework onto the feature layer. This allows us, among other things, to generate example images for a class to get a sense of what was learned. Moreover, the modular structure of the proposed model makes it possible to learn relations over multiple images at a time, as well as handle noisy labels. Our empirical evaluations show the promise of this inquiry.

In data mining, finding interesting patterns is a challenging task. Constraint-based mining is a well-known approach to this, and one for which constraint programming has been shown to be a well-suited and generic framework. Constraint dominance programming (CDP) has been proposed as an extension that can capture an even wider class of constraint-based mining problems, by allowing us to compare relations between patterns. In this paper we improve CDP with the ability to specify an incomparability condition. This allows us to overcome two major shortcomings of CDP: finding dominated solutions that must then be filtered out after search, and unnecessarily adding dominance blocking constraints between incomparable solutions. We demonstrate the efficacy of our approach by extending the problem specification language ESSENCE and implementing it in a solver-independent manner on top of the constraint modelling tool CONJURE. Our experiments on pattern mining tasks with both a CP solver and a SAT solver show that using the incomparability condition during search significantly improves the efficiency of dominance programming and reduces (and often eliminates entirely) the need for post-processing to filter dominated solutions. Publisher PDF

We investigate repeated win-lose coordination games and analyse when and how rational players can guarantee eventual coordination in such games. Our study involves both the setting with a protocol shared in advance as well as the scenario without an agreed protocol. In both cases, we focus on the case without any communication amongst the players once the particular game to be played has been revealed to them. We identify classes of coordination games in which coordination cannot be guaranteed in a single round, but can eventually be achieved in several rounds by following suitable coordination protocols. In particular, we study coordination using protocols invariant under structural symmetries of games under some natural assumptions, such as: priority hierarchies amongst players, different patience thresholds, use of focal groups, and gradual coordination by contact. We investigate repeated win-lose coordination games and analyse when and how rational players can guarantee eventual coordination in such games. Our study involves both the setting with a protocol shared in advance as well as the scenario without an agreed protocol. In both cases, we focus on the case without any communication amongst the players once the particular game to be played has been revealed to them. We identify classes of coordination games in which coordination cannot be guaranteed in a single round, but can eventually be achieved in several rounds by following suitable coordination protocols. In particular, we study coordination using protocols invariant under structural symmetries of games under some natural assumptions, such as: priority hierarchies amongst players, different patience thresholds, use of focal groups, and gradual coordination by contact.

As medical research becomes ever finer-grained, experiments require healthcare data in quantities that single countries cannot provide. Cross-jurisdictional data collection remains, however, extremely challenging due to the diverging legal, professional, linguistic, normative, and technological contexts of the participating countries. Medical data heterogeneity, in particular, is still a largely unsolved problem on the international level, due to the complexity of data combined with strict precision and data protection constraints. We propose a scalable solution based on a novel knowledge architecture and the corresponding knowledge graph integration methodology. Medical knowledge that drives the scalable integration process is divided into multiple functional layers and is maintained in a distributed manner across participating countries. We successfully applied the approach in the context of a research experiment across Scotland and Italy, and are currently adapting it within other initiatives of Europe-wide health data interoperability.

Scheduling and dispatching are critical enabling technologies in supercomputing and grid computing. In these contexts, scalability is an issue: we have to allocate and schedule upi?źto tens of thousands of tasks on tens of thousands of resources. This problem scale is out of reach for complete and centralized scheduling approaches. We propose a distributed allocation and scheduling paradigm called DARDIS that is lightweight, scalable and fully customizable in many domains. In DARDIS each task offloads to the available resources the computation of a probability index associated with each possible start time for the given task on the specific resource. The task then selects the proper resource and start time on the basis of the above probability. The scheduler can be customized with different policies to fit several objective functions like load balancing or makespan. We evaluate our approach in the domain of grids and supercomputers. We compare DARDIS with the most widely used algorithms used in these specific domains to show that this approach can reach better solutions in several cases.

Business intelligence and big data represent two different technologies within decision support systems. The present paper concerns the two concepts within the context of e-government. Thus, the purpose of the paper is to present the preliminary findings regarding publication patterns and topic coverage within the two technologies by conducting a comparative literature review. A total of 281 papers published in the years 2005–2014 were included in the analysis. A rapid increase of papers regarding big data were identified, the majority being journal papers. As regards business intelligence, researchers publish in conference proceedings to a greater extent. Further, big data journal papers are published within a broader range of journal topics compared to business intelligence journal papers. The paper concludes by pointing to further analyses that will be carried out within the 281 selected papers.