This paper presents a demonstration of how AI can be useful in the game design and development process of a modern board game. By using an artificial intelligence algorithm to play a substantial amount of matches of the Ticket to Ride board game and collecting data, we can analyze several features of the gameplay as well as of the game board. Results revealed loopholes in the game's rules and pointed towards trends in how the game is played. We are then led to the conclusion that large scale simulation utilizing artificial intelligence can offer valuable information regarding modern board games and their designs that would ordinarily be prohibitively expensive or time-consuming to discover manually.

Deep reinforcement learning (DRL) has shown incredible performance in learning various tasks to the human level. However, unlike human perception, current DRL models connect the entire low-level sensory input to the state-action values rather than exploiting the relationship between and among entities that constitute the sensory input. Because of this difference, DRL needs vast amount of experience samples to learn. In this paper, we propose a Multi-focus Attention Network (MANet) which mimics human ability to spatially abstract the low-level sensory input into multiple entities and attend to them simultaneously. The proposed method first divides the low-level input into several segments which we refer to as partial states. After this segmentation, parallel attention layers attend to the partial states relevant to solving the task. Our model estimates state-action values using these attended partial states. In our experiments, MANet attains highest scores with significantly less experience samples. Additionally, the model shows higher performance compared to the Deep Q-network and the single attention model as benchmarks. Furthermore, we extend our model to attentive communication model for performing multi-agent cooperative tasks. In multi-agent cooperative task experiments, our model shows 20% faster learning than existing state-of-the-art model.

When machine learning algorithms are used in life-critical or mission-critical applications (e.g., self driving cars, cyber security, surgical robotics), it is important to ensure that they provide some high-level correctness guarantees. We introduce a paradigm called Trusted Machine Learning (TML) with the goal of making learning techniques more trustworthy. We outline methods that show how symbolic analysis (specifi- cally parametric model checking) can be used to learn the dynamical model of a system where the learned model satis- fies correctness requirements specified in the form of temporal logic properties (e.g., safety, liveness). When a learned model does not satisfy the desired guarantees, we try two approaches: (1) Model Repair, wherein we modify a learned model directly, and (2) Data Repair, wherein we modify the data so that re-learning from the modified data will result in a trusted model. Model Repair tries to make the minimal changes to the trained model while satisfying the properties, whereas Data Repair tries to make the minimal changes to the dataset used to train the model for ensuring satisfaction of the properties. We show how the Model Repair and Data Repair problems can be solved for the case of probabilistic models, specifically Discrete-Time Markov Chains (DTMC) or Markov Decision Processes (MDP), when the desired properties are expressed in Probabilistic Computation Tree Logic (PCTL). Specifically, we outline how the parameter learning problem in the probabilistic Markov models under temporal logic constraints can be equivalently expressed as a non-linear optimization with non-linear rational constraints, by performing symbolic transformations using a parametric model checker. We illustrate the approach on two case studies: a controller for automobile lane changing, and query router for a wireless sensor network.

We present a technique to further narrow the gap between recipe-based and domain theory-based plan recognition through decompositional planning, a planning model that combines hierarchical reasoning as used in hierarchical task networks, and least-commitment refinement reasoning as used in partial-order causal link planning. We represent recipes through decompositional planning operators and use them to compile observed agent actions into an incomplete decompositional plan that represents them; this plan can then be input to a decompositional planner to identify the recognized plan-space plan. Our model thus synthesizes the heretofore disparate recipe-based and domain theory-based plan recognition variants into a unified knowledge representation and reasoning model.

Domain adaptation deals with adapting classifiers trained on data from a source distribution, to work effectively on data from a target distribution. In this paper, we introduce the Nonlinear Embedding Transform (NET) for unsupervised domain adaptation. The NET reduces cross-domain disparity through nonlinear domain alignment. It also embeds the domain-aligned data such that similar data points are clustered together. This results in enhanced classification. To determine the parameters in the NET model (and in other unsupervised domain adaptation models), we introduce a validation procedure by sampling source data points that are similar in distribution to the target data. We test the NET and the validation procedure using popular image datasets and compare the classification results across competitive procedures for unsupervised domain adaptation.

Knowledge Graph (KG) plays a crucial role in many modern applications. Nevertheless, constructing KG from unstructured text is a challenging problem due to its nature. Consequently, many approaches propose to transform unstructured text to structured text in order to create a KG. Such approaches cannot yet provide reasonable results for mapping an extracted predicate to its identical predicate in another KG. Predicate mapping is an essential procedure because it can reduce the heterogeneity problem and increase searchability over a KG. In this paper, we propose T2KG system, an end-to-end system with keeping such problem into consideration. In the system, a hybrid combination of a rule-based approach and a similarity-based approach is presented for mapping a predicate to its identical predicate in a KG. Based on preliminary experimental results, the hybrid approach improves the recall by 10.02% and the F-measure by 6.56% without reducing the precision in the predicate mapping task. Furthermore, although the KG creation is conducted in open domains, the system still achieves approximately 50% of F-measure for generating triples in the KG creation task.

Distributed knowledge representation learning (KRL) methods encode both entities and relations in knowledge graphs (KG) in a lower-dimensional semantic space, which model relatively dense knowledge graphs well and greatly improve the performance of knowledge graph completion and knowledge reasoning. However, existing KRL methods including Trans(E, H, R, D and Sparse) hardly obtain comparative performances on sparse KGs where most of entities and relations have very low frequencies. Furthermore, all existing methods target at KRL on one knowledge graph independently. The embeddings of different KGs are independent with each other. In this paper, we propose a novel cross-knowledge-graph (cross-KG) KRL method which learns embeddings for two different KGs simultaneously. Through projecting semantic related entities and relations in two KGs to a uniform semantic space, our method could learn better embeddings for sparse KGs by incorporating information from another relatively larger and denser KG. The learned embeddings are also helpful for downstream cross-KGs or cross-linguals tasks like ontology alignment. The experiment results show that our method could significantly outperform corresponding baseline methods on knowledge graph completion on single KG and cross-KG entity prediction and mapping tasks.

To ensure we do not forget relevant aspects of AI, we The field of Artificial Intelligence (AI) is fertile: it is at the present some key works which have already focused on same time the root of the dreams and deceptions of many defining (artificial) intelligence in Section 2. We then highlight people, a common feature in science fiction, and various the potential lack of cross-fertilisation they may be subject technical projects in many domains of application. Although to in Section 3 and consider the definition of human we may appreciate the rich emotions and ideas brought by expertise to draw a definition of human intelligence in Section a concept such as AI, some people are seriously working on 4. Next, we generalise these definitions to cover also artificial it in an attempt to produce autonomous agents able to meet agents in Section 5 and provide more details about the the various needs of different users. These projects, however, domain-generic data and processes of our definition of intelligence have faced several troubles and unfulfilled promises in in Section 6. We rely further on the expertise field in the history of the field, leading to shortenings of funding Section 7 by describing three kinds of measures of expertise, and years of research efforts lost (Franklin 2014). Despite mapping them to existing measures of intelligence, and suggesting the presence of "intrepid researchers" to advance the field, directions to investigate. Finally, Section 8 expands from an industrial point of view such projects were abandoned the discussion to a novel conception of the field of AI as a and considered as failures.

In this paper we describe our NLP supported programming-by-demonstration approach to high-level robot programming that allows users to generate skills and robot program primitives for later refinement and re-use. Our ideas incorporate the identification of common user strategies (interaction patterns) in the programming process, which can be exploited to support a human user in establishing common ground with a robotic system. We have evaluated a prototype implementation of this approach in a user study and use observations from this study to define further research efforts, which we discuss in this short paper.

The problem of ordering documents in a large collection into a sequence that is efficient for learning (both human and machine) is of high practical significance, but has not yet been well-formulated. We formulate this problem as mining a maximally informative simple sequence of documents. The mined sequence should be maximally informative in the sense that the reader learns quickly by reading only a few documents, and it should be simple so that the reader is not overwhelmed while trying to learn the content. The task can be posed as: Given that a reader wishes to read (at most) k documents, which documents should be selected from the repository and in what order, so as to provide maximum information. We present the WikiSeq algorithm for this purpose. We also design a metric based on information-gain to help objectively evaluate WikiSeq, and conduct experiments to compare with indicative baselines. Finally, we provide case-studies to subjectively illustrate WikiSeq’s merits.