This paper investigates the problem of cross-lingual transfer parsing, aiming at inducing dependency parsers for low-resource languages while using only training data from a resource-rich language (e.g., English). Existing model transfer approaches typically don't include lexical features, which are not transferable across languages. In this paper, we bridge the lexical feature gap by using distributed feature representations and their composition. We provide two algorithms for inducing cross-lingual distributed representations of words, which map vocabularies from two different languages into a common vector space. Consequently, both lexical features and non-lexical features can be used in our model for cross-lingual transfer. Furthermore, our framework is flexible enough to incorporate additional useful features such as cross-lingual word clusters. Our combined contributions achieve an average relative error reduction of 10.9% in labeled attachment score as compared with the delexicalized parser, trained on English universal treebank and transferred to three other languages. It also significantly outperforms state-of-the-art delexicalized models augmented with projected cluster features on identical data. Finally, we demonstrate that our models can be further boosted with minimal supervision (e.g., 100 annotated sentences) from target languages, which is of great significance for practical usage.

As a reminder, all participants' lines will be in the listen-only mode. There will be an opportunity for you to ask questions after the presentation concludes. I would now like to hand the conference over to Mr. Aravind Viswanathan. Thank you and over to you, sir. We will begin the call with business highlights and overview by Abid, the Chief Executive Officer and Member of the Board, followed by the financial overview by our CFO, Jatin Dalal. Afterwards, the operator will open the bridge for Q&A with our management team. Before Abid starts, let me draw your attention to the fact that during this call, we may make certain forward-looking statements within the meaning of Private Securities Litigation Reform Act 1995. These statements are based on management's current expectations and are associated with uncertainties and risks, which may cause the actual results to differ materially from those expected. The uncertainties and risk factors are being explained in our detailed filings with the SEC. Wipro does not undertake any obligation to update the forward-looking statements to reflect events and circumstances after the date of filing thereof. The conference call will be archived and the transcript will be available on our website. Ladies and gentlemen, let me now hand it over to Mr. Abid. Today is the first opportunity for me to interact with all of you since I've taken over as the Chief Executive Officer of Wipro, and it's a special moment for me. While I will speak about the performance of our full quarter and the full fiscal year, I thought I will take this opportunity to begin by speaking about our ambition, our strategy and how we are going to execute this strategy. Since I got announced within two days, I was able to define and announce my structure and I had already preselected my leadership team which I announced on 6th of January, effective February 1. Over the past 80 days after I have taken over as CEO, I've had the opportunity to go around the globe and meet about 70 of our top 100 clients. And both with my leadership team and with the customers, I've had the opportunity to validate the strategy that we have been working on and this gives me a high level of confidence on the relevance of our overall strategy. Our ambition is to double our revenues to 15 billion by fiscal 2020 with a 23% operating margin.

Omoju Miller, a self-described futurist (someone who studies the future's possibilities), enjoys picturing tomorrow. As a Nigerian woman who settled in the Bay Area, she's already torn down historical barriers to work as a software engineer in Silicon Valley, a white man's world. But in envisioning a new society, Miller isn't thinking only of contemporary struggles; she's pondering what humanity will need next. Take one of her projects: Hiphopathy, where she's using machine learning to parse rappers' metaphorical language, in the hopes of teaching a computer to think conceptually, developing, in the process, a form of artificial intelligence. Recently, NationSwell spoke with Miller about true visionaries that inspire her and the lessons we can all take away from their avant-garde thinking.

This time last year Steve Wozniak was sounding a cautionary note about the future of Artificial Intelligence (AI), warning that computers would one day take over from humans and joking that we might even end up as their pets. In a recent interview with Australia's ABC TV's Lateline the engineering genius appeared more sanguine about the future of self-aware, super-intelligent Artificial Intelligence and much more concerned with the real world killer robots that are all but with us: Lethal Autonomous Weapon Systems (LAWS). The Apple co-founder maintains that human-level Artificial Intelligence won't happen for "a very long time": It might take 200 years before they are really fully able to operate all of their needs in the world, until then they're going to need human beings … I'm not really worried at all. It's very scary to make autonomous weapons that are just following some programmed set of instructions … even when you're driving a car there is no one set of rules … if a lane is closed off you have to do something against the rules … I don't think it's a good idea at all. I don't think we can really stop it.

Which industries will come to the fore in the next decade, and beyond, and become hubs of innovation? According to former State Department official Alec Ross, they won't be the industries that have dominated technology thus far. Instead, artificial intelligence (AI), genomics and robotics will lead the way. On Tuesday, the Italian Embassy in Washington, D.C., held an event to discuss Ross' recently published book, The Industries of the Future. He expounded on the book's themes and highlighted what it will take for individuals, companies and countries to harness the changes that he sees coming to the global economy.

Niall Dunne is the Chief Sustainability Officer for BT, working with BT's Chief Executive, Chairman and executive management team to bring the company's purpose, to use the power of communications to make a better world, to life. Before joining BT in 2011, Niall was Managing Director in Europe, the Middle East and Africa (EMEA) at Saatchi & Saatchi. Prior to that, Dunne was an executive at Accenture, where he helped establish the company's climate change and sustainability practice. Dunne has written and spoken about the power of communications to tackle major social, environmental and economic problems. Niall was vice chair of the WEF's Global Agenda Council on Sustainable Consumption 2012-14 and joined the WEF Global Agenda Council on Climate Change in 2014.

We consider the problem of obtaining coherence in a propositional knowledge base using techniques from Belief Change. Our motivation comes from the field of formal ontologies where coherence is interpreted to mean that a concept name has to be satisfiable. In the propositional case we consider here, this translates to a propositional formula being satisfiable. We define belief change operators in a framework of nonmonotonic preferential reasoning.We show how the introduction of defeasible information using contraction operators can be an effective means for obtaining coherence.

Recent years have seen increasing interest in AI from outside the AI community. This is partly due to applications based on AI that have been used in real-world domains, for example, the successful deployment of game theory-based decision aids in security domains. This paper describes our teaching approach for introducing the AI concepts underlying security games to diverse audiences. We adapted a game-based research platform that served as a testbed for recent research advances in computational game theory into a set of interactive role-playing games. We guided learners in playing these games as part of our teaching strategy, which also included didactic instruction and interactive exercises on broader AI topics. We describe our experience in applying this teaching approach to diverse audiences, including students of an urban public high school, university undergraduates, and security domain experts who protect wildlife. We evaluate our approach based on results from the games and participant surveys.

Robotic tactile recognition aims at identifying target objects or environments from tactile sensory readings. The advancement of unsupervised feature learning and biological tactile sensing inspire us proposing the model of 3T-RTCN that performs spatio-temporal feature representation and fusion for tactile recognition. It decomposes tactile data into spatial and temporal threads, and incorporates the strength of randomized tiling convolutional networks. Experimental evaluations show that it outperforms some state-of-the-art methods with a large margin regarding recognition accuracy, robustness, and fault-tolerance; we also achieve an order-of-magnitude speedup over equivalent networks with pretraining and finetuning. Practical suggestions and hints are summarized in the end for effectively handling the tactile data.

Entity Set Expansion (ESE) and Attribute Extraction (AE) are usually treated as two separate tasks in Information Extraction (IE). However, the two tasks are tightly coupled, and each task can benefit significantly from the other by leveraging the inherent relationship between entities and attributes. That is, 1) an attribute is important if it is shared by many typical entities of a class; 2) an entity is typical if it owns many important attributes of a class. Based on this observation, we propose a joint model for ESE and AE, which models the inherent relationship between entities and attributes as a graph. Then a graph reinforcement algorithm is proposed to jointly mine entities and attributes of a specific class. Experimental results demonstrate the superiority of our method for discovering both new entities and new attributes.