We study the problem of reward shaping to accelerate the training process of a reinforcement learning agent. Existing works have considered a number of different reward shaping formulations; however, they either require external domain knowledge or fail in environments with extremely sparse rewards.

While the expressive power and computational capabilities of graph neural networks (GNNs) have been theoretically studied, their optimization and learning dynamics, in general, remain largely unexplored. Our study undertakes the Graph Attention Network (GAT), a popular GNN architecture in which a node's neighborhood aggregation is weighted by parameterized attention coefficients. We derive a conservation law of GAT gradient flow dynamics, which explains why a high portion of parameters in GATs with standard initialization struggle to change during training. This effect is amplified in deeper GATs, which perform significantly worse than their shallow counterparts. To alleviate this problem, we devise an initialization scheme that balances the GAT network. Our approach i) allows more effective propagation of gradients and in turn enables trainability of deeper networks, and ii) attains a considerable speedup in training and convergence time in comparison to the standard initialization. Our main theorem serves as a stepping stone to studying the learning dynamics of positive homogeneous models with attention mechanisms.

We study the problem of reward shaping to accelerate the training process of a reinforcement learning agent. Existing works have considered a number of different reward shaping formulations; however, they either require external domain knowledge or fail in environments with extremely sparse rewards.

Automatic detection of voice pathology enables objective assessment and earlier intervention for the diagnosis. This study provides a systematic analysis of glottal source features and investigates their effectiveness in voice pathology detection. Glottal source features are extracted using glottal flows estimated with the quasi-closed phase (QCP) glottal inverse filtering method, using approximate glottal source signals computed with the zero frequency filtering (ZFF) method, and using acoustic voice signals directly. In addition, we propose to derive mel-frequency cepstral coefficients (MFCCs) from the glottal source waveforms computed by QCP and ZFF to effectively capture the variations in glottal source spectra of pathological voice. Experiments were carried out using two databases, the Hospital Universitario Principe de Asturias (HUPA) database and the Saarbrucken Voice Disorders (SVD) database. Analysis of features revealed that the glottal source contains information that discriminates normal and pathological voice. Pathology detection experiments were carried out using support vector machine (SVM). From the detection experiments it was observed that the performance achieved with the studied glottal source features is comparable or better than that of conventional MFCCs and perceptual linear prediction (PLP) features. The best detection performance was achieved when the glottal source features were combined with the conventional MFCCs and PLP features, which indicates the complementary nature of the features.

Both conceptual modeling and machine learning have long been recognized as important areas of research. With the increasing emphasis on digitizing and processing large amounts of data for business and other applications, it would be helpful to consider how these areas of research can complement each other. To understand how they can be paired, we provide an overview of machine learning foundations and development cycle. We then examine how conceptual modeling can be applied to machine learning and propose a framework for incorporating conceptual modeling into data science projects. The framework is illustrated by applying it to a healthcare application. For the inverse pairing, machine learning can impact conceptual modeling through text and rule mining, as well as knowledge graphs. The pairing of conceptual modeling and machine learning in this this way should help lay the foundations for future research.

The Confederation of Laboratories for Artificial Intelligence Research in Europe (CLAIRE) has announced the launch of four new offices. These will be located in Zürich, Oslo, Paris and Brussels and will complement the existing offices in The Hague, Prague, Rome and Saarbrücken. There will be a launch roadshow beginning on 26 May which will take the form of a series of virtual events. For each event, the offices will invite national and international speakers, including government representatives and panellists from various areas of AI, to talk about topics related to the designated focus area of the respective office. Each CLAIRE Office has its own focus in supporting work on current topics around European and human-centred AI research, AI for good and AI for all.

Almost every word has more than one meaning. Modern search engines solve this problem using knowledge bases. Yago was one of the first knowledge bases, developed by scientists at the Max Planck Institute for Informatics in Saarbrücken and the Télécom ParisTech in Paris. Last week, the researchers received an award for their work on Yago from the most important scientific journal in the field of artificial intelligence. Today, they are releasing Yago's source code.

The aim of their collaboration is to achieve machine-based translation between the languages of the European Union so that comprehensible texts are achieved for as many language combinations as possible. Two of the EU-funded research projects are being led by the Saarbrücken computer linguist Josef van Genabith. Anyone who wants to learn Finnish has to be prepared to deal with a complex grammar that includes fifteen different cases. The grammatical cases are marked in part by appending syllables to nouns resulting in a dizzying array of word forms and expressive possibilities. "Teaching a computer to understand all these grammatical nuances and to translate them correctly into another language is exceptionally difficult," says Josef van Genabith, Professor of Translation-Oriented Language Technologies at Saarland University and a Scientific Director at the German Research Center for Artificial Intelligence (DFKI).

The dinner speech at the Fischerhuette was given by Jörg Siekmann (University of Saarbrücken). The poster session, consisting of 18 posters and demos, was jointly organized as a get-together with the Berlin Semantic Web Meetup. At the session, wine, beer, and finger food were provided in the greenhouses of the Computer Science Department at The Thirty-First AAAI Conference on Artificial Intelligence the Freie Universität Berlin. The organizers also used (AAAI-17) and the Twenty-Ninth Conference on Innovative this unique opportunity to hold a joint public Applications of Artificial Intelligence (IAAI-17), will be RuleML and RR business meeting as well as an invited held in New Orleans, Louisiana, USA, during the mid-January dinner with all chairs, and invited keynote speakers to mid-February timeframe. AAAI-17 August 1, a boat sightseeing tour from lake Wannsee will arrive in New Orleans just prior to Mardi Gras and festivities to the Reichstag on Sunday, August 2, the CADE exhibitions will already be underway.

This article summarizes the results of the 6-7 July Workshop on Human Language Technology and Knowledge Management held in Toulouse, France. It describes invited keynotes, presentations, and results of brainstorming sessions to create a technology road map for this important area. The group also articulated grand challenges in human language technology and solutions to these challenges that could benefit facilities for knowledge discovery, access, and exploitation.