There has been a surge in research in multimodal machine translation (MMT), where additional modalities such as images are used to improve translation quality of textual systems. A particular use for such multimodal systems is the task of simultaneous machine translation, where visual context has been shown to complement the partial information provided by the source sentence, especially in the early phases of translation. In this paper, we propose the first Transformer-based simultaneous MMT architecture, which has not been previously explored in simultaneous translation. Additionally, we extend this model with an auxiliary supervision signal that guides the visual attention mechanism using labelled phrase-region alignments. We perform comprehensive experiments on three language directions and conduct thorough quantitative and qualitative analyses using both automatic metrics and manual inspection. Our results show that (i) supervised visual attention consistently improves the translation quality of the simultaneous MMT models, and (ii) fine-tuning the MMT with supervision loss enabled leads to better performance than training the MMT from scratch. Compared to the state-of-the-art, our proposed model achieves improvements of up to 2.3 BLEU and 3.5 METEOR points.

Discrete-action algorithms have been central to numerous recent successes of deep reinforcement learning. However, applying these algorithms to high-dimensional action tasks requires tackling the combinatorial increase of the number of possible actions with the number of action dimensions. This problem is further exacerbated for continuous-action tasks that require fine control of actions via discretization. In this paper, we propose a novel neural architecture featuring a shared decision module followed by several network branches, one for each action dimension. This approach achieves a linear increase of the number of network outputs with the number of degrees of freedom by allowing a level of independence for each individual action dimension. To illustrate the approach, we present a novel agent, called Branching Dueling Q-Network (BDQ), as a branching variant of the Dueling Double Deep Q-Network (Dueling DDQN). We evaluate the performance of our agent on a set of challenging continuous control tasks. The empirical results show that the proposed agent scales gracefully to environments with increasing action dimensionality and indicate the significance of the shared decision module in coordination of the distributed action branches. Furthermore, we show that the proposed agent performs competitively against a state-of-the-art continuous control algorithm, Deep Deterministic Policy Gradient (DDPG).

There is growing need for robots that can interact with people in everyday situations. For service robots, it is not reasonable to assume that one can pre-program all object categories. Instead, apart from learning from a batch of labelled training data, robots should continuously update and learn new object categories while working in the environment. This paper proposes a cognitive architecture designed to create a concurrent 3D object category learning and recognition in an interactive and open-ended manner. In particular, this cognitive architecture provides automatic perception capabilities that will allow robots to detect objects in highly crowded scenes and learn new object categories from the set of accumulated experiences in an incremental and open-ended way. Moreover, it supports constructing the full model of an unknown object in an on-line manner and predicting next best view for improving object detection and manipulation performance. We provide extensive experimental results demonstrating system performance in terms of recognition, scalability, next-best-view prediction and real-world robotic applications.

We study the complexity of bribery in a network-based rating system, where individuals are connected in a social network and an attacker, typically a service provider, can influence their rating and increase the overall profit. We derive a number of algorithmic properties of this framework, in particular we show that establishing the existence of an optimal manipulation strategy for the attacker is NP-complete, even with full knowledge of the underlying network structure.

The study of properties of gradual evaluation methods in argumentation has received increasing attention in recent years, with studies devoted to various classes of frameworks/methods leading to conceptually similar but formally distinct properties in different contexts. In this paper we provide a systematic analysis for this research landscape by making three main contributions. First, we identify groups of conceptually related properties in the literature, which can be regarded as based on common patterns and, using these patterns, we evidence that many further properties can be considered. Then, we provide a simplifying and unifying perspective for these properties by showing that they are all implied by the parametric principles of (either strict or non-strict) balance and monotonicity. Finally, we show that (instances of) these principles are satisfied by several quantitative argumentation formalisms in the literature, thus confirming their general validity and their utility to support a compact, yet comprehensive, analysis of properties of gradual argumentation.

The graph Laplacian plays key roles in information processing of relational data, and has analogies with the Laplacian in differential geometry. In this paper, we generalize the analogy between graph Laplacian and differential geometry to the hypergraph setting, and propose a novel hypergraph p-Laplacian. Unlike the existing two-node graph Laplacians, this generalization makes it possible to analyze hypergraphs, where the edges are allowed to connect any number of nodes. Moreover, we propose a semi-supervised learning method based on the proposed hypergraph p-Laplacian, and formalize them as the analogue to the Dirichlet problem, which often appears in physics. We further explore theoretical connections to normalized hypergraph cut on a hypergraph, and propose normalized cut corresponding to hypergraph p-Laplacian. The proposed p-Laplacian is shown to outperform standard hypergraph Laplacians in the experiment on a hypergraph semi-supervised learning and normalized cut setting.

We revisit the problem of robust principal component analysis with features acting as prior side information. To this aim, a novel, elegant, non-convex optimization approach is proposed to decompose a given observation matrix into a low-rank core and the corresponding sparse residual. Rigorous theoretical analysis of the proposed algorithm results in exact recovery guarantees with low computational complexity. Aptly designed synthetic experiments demonstrate that our method is the first to wholly harness the power of non-convexity over convexity in terms of both recoverability and speed. That is, the proposed non-convex approach is more accurate and faster compared to the best available algorithms for the problem under study. Two real-world applications, namely image classification and face denoising further exemplify the practical superiority of the proposed method.

The increase of connectivity and the impact it has in every day life is raising new and existing security problems that are becoming important for social good. We introduce two particular problems: cyber attack attribution and regulatory data sharing. For both problems, decisions about which rules to apply, should be taken under incomplete and context dependent information. The solution we propose is based on argumentation reasoning, that is a well suited technique for implementing decision making mechanisms under conflicting and incomplete information. Our proposal permits us to identify the attacker of a cyber attack and decide the regulation rule that should be used while using and sharing data. We illustrate our solution through concrete examples.

Participatory sensing (PS) has gained significant attention as a crowdsourcing methodology that allows ordinary citizens (non-expert contributors) to collect data using low-cost mobile devices. In particular, it has been useful in the collection of environmental data. However, current PS applications suffer from two problems. First, they do not coordinate the measurements taken by their users, which is required to maximise system efficiency. Second, they are vulnerable to malicious behaviour. In this context, we propose a novel algorithm that simultaneously addresses both of these problems. Specifically, we use heteroskedastic Gaussian Processes to incorporate users' trustworthiness into a Bayesian spatio-temporal regression model. The model is trained with measurements taken by participants, thus it is able to estimate the value of the phenomenon at any spatio-temporal location of interest and also learn the level of trustworthiness of each user. Given this model, the coordination system is able to make informed decisions concerning when, where and who should take measurements over a period of time. We empirically evaluate our algorithm on a real-world human mobility and air quality dataset, where malicious behaviour is synthetically produced, and show that our algorithm outperforms the current state of the art by up to 60.4% in terms of RMSE while having a reasonable runtime.

We define a class of parameterised infinite state multi-agent systems (MAS) that is unbounded in both the number of agents composing the system and the domain of the variables encoding the agents. We analyse their verification problem by combining and extending existing techniques in parameterised model checking with predicate abstraction procedures. The resulting methodology addresses both forms of unboundedness and provides a technique for verifying unbounded MAS defined on infinite-state variables. We illustrate the effectiveness of the technique on an infinite-domain variant of an unbounded version of the train-gate-controller.