Implicit discourse relation recognition is a challenging task as the relation prediction without explicit connectives in discourse parsing needs understanding of text spans and cannot be easily derived from surface features from the input sentence pairs. Thus, properly representing the text is very crucial to this task. In this paper, we propose a model augmented with different grained text representations, including character, subword, word, sentence, and sentence pair levels. The proposed deeper model is evaluated on the benchmark treebank and achieves state-of-the-art accuracy with greater than 48% in 11-way and $F_1$ score greater than 50% in 4-way classifications for the first time according to our best knowledge.

Abstract-- Autonomous systems will play an essential role in many applications across diverse domains including space, marine, air, field, road, and service robotics. They will assist us in our daily routines and perform dangerous, dirty and dull tasks. However, enabling robotic systems to perform autonomously in complex, real-world scenarios over extended time periods (i.e. Some of these have been investigated by sub-disciplines of Artificial Intelligence (AI) including navigation & mapping, perception, knowledge representation & reasoning, planning, interaction, and learning. The different sub-disciplines have developed techniques that, when re-integrated within an autonomous system, can enable robots to operate effectively in complex, long-term scenarios. In this paper, we survey and discuss AI techniques as'enablers' for long-term robot autonomy, current progress in integrating these techniques within long-running robotic systems, and the future challenges and opportunities for AI in long-term autonomy. I. INTRODUCTION Robot technology has improved tremendously over the last decade. Consequently, autonomous robot systems have been able to operate in increasingly complex environments and for increasingly long periods of time, i.e. weeks, months, or years. When a fully modelled robot is deployed in a completely known, static environment, the challenge of long-term autonomy (LTA) reduces to one of robustness, i.e. enabling the robot to remain operational for as long as possible. Without these simplifying assumptions autonomous robots face a number of interrelated challenges. The first refers to the application requirements, e.g., the robot platform (hardware and software), environment and tasks to be performed.

We introduce "Talk The Walk", the first large-scale dialogue dataset grounded in action and perception. The task involves two agents (a "guide" and a "tourist") that communicate via natural language in order to achieve a common goal: having the tourist navigate to a given target location. The task and dataset, which are described in detail, are challenging and their full solution is an open problem that we pose to the community. We (i) focus on the task of tourist localization and develop the novel Masked Attention for Spatial Convolutions (MASC) mechanism that allows for grounding tourist utterances into the guide's map, (ii) show it yields significant improvements for both emergent and natural language communication, and (iii) using this method, we establish non-trivial baselines on the full task.

Deep learning with neural networks is applied by an increasing number of people outside of classic research environments, due to the vast success of the methodology on a wide range of machine perception tasks. While this interest is fueled by beautiful success stories, practical work in deep learning on novel tasks without existing baselines remains challenging. This paper explores the specific challenges arising in the realm of real world tasks, based on case studies from research \& development in conjunction with industry, and extracts lessons learned from them. It thus fills a gap between the publication of latest algorithmic and methodical developments, and the usually omitted nitty-gritty of how to make them work. Specifically, we give insight into deep learning projects on face matching, print media monitoring, industrial quality control, music scanning, strategy game playing, and automated machine learning, thereby providing best practices for deep learning in practice.

We introduce a new entity typing task: given a sentence with an entity mention, the goal is to predict a set of free-form phrases (e.g. skyscraper, songwriter, or criminal) that describe appropriate types for the target entity. This formulation allows us to use a new type of distant supervision at large scale: head words, which indicate the type of the noun phrases they appear in. We show that these ultra-fine types can be crowd-sourced, and introduce new evaluation sets that are much more diverse and fine-grained than existing benchmarks. We present a model that can predict open types, and is trained using a multitask objective that pools our new head-word supervision with prior supervision from entity linking. Experimental results demonstrate that our model is effective in predicting entity types at varying granularity; it achieves state of the art performance on an existing fine-grained entity typing benchmark, and sets baselines for our newly-introduced datasets. Our data and model can be downloaded from: http://nlp.cs.washington.edu/entity_type

We introduce a new generative model for human planning under the Bayesian Inverse Reinforcement Learning (BIRL) framework which takes into account the fact that humans often plan using hierarchical strategies. We describe the Bayesian Inverse Hierarchical RL (BIHRL) algorithm for inferring the values of hierarchical planners, and use an illustrative toy model to show that BIHRL retains accuracy where standard BIRL fails. Furthermore, BIHRL is able to accurately predict the goals of `Wikispeedia' game players, with inclusion of hierarchical structure in the model resulting in a large boost in accuracy. We show that BIHRL is able to significantly outperform BIRL even when we only have a weak prior on the hierarchical structure of the plans available to the agent, and discuss the significant challenges that remain for scaling up this framework to more realistic settings.

Mathematically, ternary coding is more efficient than binary coding. It is little used in computation because technology for binary processing is already established and the implementation of ternary coding is more complicated, but remains relevant in algorithms that use decision trees and in communications. In this paper we present a new comparison of binary and ternary coding and their relative efficiencies are computed both for number representation and decision trees. The implications of our inability to use optimal representation through mathematics or logic are examined. Apart from considerations of representation efficiency, ternary coding appears preferable to binary coding in classification of many real-world problems of artificial intelligence (AI) and medicine. We examine the problem of identifying appropriate three classes for domain-specific applications. Keywords: optimal coding, decision trees, ternary logic, artificial intelligence Introduction The problem of optimal coding of numbers has been examined by many scholars (e.g.

The astonishing thing about stem cells is that they can be coaxed, in the laboratory, into becoming nearly any kind of cell--from bone marrow to heart muscle. That remarkable capability has for years kept scientists busy tinkering with stem cells and injecting them into animal models in an attempt to grow and replace damaged tissue. Such scientists have received a ton of attention in that line of work. But there's a smaller group of researchers working, to far less fanfare, on a different part of the stem cell challenge: how to deliver those cells to the body's hard-to-reach places. Researchers typically deliver stem cells via injection--a needle.

IF there are two truths that are universally acknowledged, they are that western populations are ageing and that more jobs are likely to be automated. That is the focus of a new report called "The Twin Threats of Aging and Automation", a collaboration between Marsh & McLennan's Global Risk Centre, Mercer, and Oliver Wyman. The study looks at 15 countries and concludes that Asian workers are most at risk (see chart). By 2050, the UN estimates that more than a third of the world's population will be over 50, up from less than 16% in 1950. As a proportion of the working-age population, those aged between 50 and 64 already make up more than 30% of the workforce in Canada, Germany, Italy and Japan.

A self driving robotic racing car is taking on the world's best human drivers at the Goodwood Festival of Speed. The Roborace car, which is powered by four 135kW electric motors and uses an artificial intelligence driver, is shown in a practice run driving up the event's 1.16-mile hillclimb course, famed for its tight turns,hay bales, flint walls and forests. It has previously raced city circuits around the world as part of the Formula E race series - and later this week will compete against times set by human drivers on the famous course. The Roborace car, which is powered by four 135kW electric motors and uses an artificial intelligence driver, is shown in a practice run driving up the event's 1.16-mile hillclimb course, famed for its tight turns, hay bales, flint walls and forests. 'We are excited that the Duke of Richmond [FoS founder] has invited us to make history at Goodwood as we attempt the first ever fully - and truly - autonomous uphill climb using only artificial intelligence,' said Lucas di Grassi, Roborace CEO.