There are currently few datasets appropriate for training and evaluating models for non-goal-oriented dialogue systems (chatbots); and equally problematic, there is currently no standard procedure for evaluating such models beyond the classic Turing test. The aim of our competition is therefore to establish a concrete scenario for testing chatbots that aim to engage humans, and become a standard evaluation tool in order to make such systems directly comparable. This is the second Conversational Intelligence (ConvAI) Challenge. The previous one was conducted under the scope of NIPS 2017 Competitions track. The winning entry will receive $20,000 in Mechanical Turk funding – in order to encourage further data collection for dialogue research.

The United States could lose its economic and political standing in the world if it doesn't develop a comprehensive, high-priority plan for artificial intelligence. That's the stark warning from a new report issued today by the Center for a New American Security (CNAS), a think tank in Washington, DC. World order: Other nations, including China, India, France, and the UK, have embraced artificial intelligence by announcing major national initiatives. Many experts recommend that the US follow suit, but the Trump White House has taken a surprisingly hands-off approach. Falling behind in AI development and implementation would present a risk for U.S. global economic and military leadership.

Of all corporate functions, sales by its very nature is surely the most people-focused. While it may no longer involve quite as much face-to-face interaction as it once did, selling has remained emphatically a job for people rather than machines. However, artificial intelligence (AI) and machine-learning are already starting to make major inroads into the sales process, adding an extra dimension to everything from marketing automation to customer relationship management. According to Salesforce Research, high-performing teams are at least twice as likely to be using intelligent sales technologies such as artificial intelligence, sentiment analysis, next-step analysis and deep-learning. So, what further changes in the sales environment can we expect to see over the coming years?

A manuscript written by Ada Lovelace, who's considered by many to be the first computer programmer, was just sold at auction for more than $125,000, the Guardian reports. A first edition and just one of six known copies of the book, it contains Lovelace's translation of a paper written by Italian mathematician Luigi Menabrea that describes Charles Babbage's plans for his "Analytical Engine" computing machine. The manuscript also contains copious explanatory notes and observations from Lovelace as well as a formula for calculating Bernoulli numbers that has been called the world's first computer program. Lovelace was the daughter of Lord Byron and her interest and aptitude for mathematics led to a years-long friendship and intellectual partnership with inventor Charles Babbage. "Until recently, her story of mathematical excellence was a lesser known one," author Kate Pankhurst told the Guardian.

Civil liberties group Big Brother Watch has launched a legal challenge against the use of automatic facial recognition technology by London's Metropolitan Police force. The privacy campaigners described the Met's "China-style" facial recognition system, which uses AI software to match people's faces to a criminal database, as "dangerously authoritarian." "Facial recognition is the latest Orwellian mass surveillance tool to be lawlessly rolled out by the state," Big Brother Watch writes on the campaign website. "These real-time facial recognition cameras are biometric checkpoints, identifying members of the public without their knowledge. Police have begun feeding secret watchlists to the cameras, containing not only criminals but suspects, protesters, football fans and innocent people with mental health problems."

RNNs and their variants have been widely adopted for image captioning. In RNNs, the production of a caption is driven by a sequence of latent states. Existing captioning models usually represent latent states as vectors, taking this practice for granted. We rethink this choice and study an alternative formulation, namely using two-dimensional maps to encode latent states. This is motivated by the curiosity about a question: how the spatial structures in the latent states affect the resultant captions? Our study on MSCOCO and Flickr30k leads to two significant observations. First, the formulation with 2D states is generally more effective in captioning, consistently achieving higher performance with comparable parameter sizes. Second, 2D states preserve spatial locality. Taking advantage of this, we visually reveal the internal dynamics in the process of caption generation, as well as the connections between input visual domain and output linguistic domain.

Named Entity Recognition (NER) is a crucial part of various Natural Language Processing (NLP) tasks like entity linking, relation extraction, machine reading and ultimately Question Answering (QA). With the recent rise of neural networks, much emphasis has been put on high-resource languages like English or Chinese leading to fast advancements of many foundational tasks, in particular NER which in many areas reaches near-human performance for these languages [1], [2]. However, for other, less-resource languages like German, their neural NER counterparts did not attract similar attention from the deep learning community, leading to lower performance by a margin of up to 11% F-score. In this paper, we look for the reasons and take steps towards solving them. By example of German we bridge the current gap between the performance of neural NER for different languages and bring the performance to a new state-of-theart. We report evidence that the inferior quality of German text data and its small size are the major reasons for the observed lack of progress. To tackle this problem, we use a larger corpus for training the foundational word embeddings, namely Leipzig40 [3] (including the whole German Wikipedia till 2016) combined with the WMT 2010 German monolingual training data [4], and contrast its use with the COW corpus [5], the largest collection of German texts extracted from web documents with over 617 Mio.

In this paper, we propose a new framework to study the generalization property of classifier chains trained over observations associated with multiple and interdependent class labels. The results are based on large deviation inequalities for Lipschitz functions of weakly dependent sequences proposed by Rio in 2000. We believe that the resulting generalization error bound brings many advantages and could be adapted to other frameworks that consider interdependent outputs. First, it explicitly exhibits the dependencies between class labels. Secondly, it provides insights of the effect of the order of the chain on the algorithm generalization performances. Finally, the two dependency coefficients that appear in the bound could also be used to design new strategies to decide the order of the chain.

Neural Turing Machines (NTMs) are an instance of Memory Augmented Neural Networks, a new class of recurrent neural networks which decouple computation from memory by introducing an external memory unit. NTMs have demonstrated superior performance over Long Short-Term Memory Cells in several sequence learning tasks. A number of open source implementations of NTMs exist but are unstable during training and/or fail to replicate the reported performance of NTMs. This paper presents the details of our successful implementation of a NTM. Our implementation learns to solve three sequential learning tasks from the original NTM paper. We find that the choice of memory contents initialization scheme is crucial in successfully implementing a NTM.

Selective clustering annotated using modes of projections (SCAMP) is a new clustering algorithm for data in $\mathbb{R}^p$. SCAMP is motivated from the point of view of non-parametric mixture modeling. Rather than maximizing a classification likelihood to determine cluster assignments, SCAMP casts clustering as a search and selection problem. One consequence of this problem formulation is that the number of clusters is $\textbf{not}$ a SCAMP tuning parameter. The search phase of SCAMP consists of finding sub-collections of the data matrix, called candidate clusters, that obey shape constraints along each coordinate projection. An extension of the dip test of Hartigan and Hartigan (1985) is developed to assist the search. Selection occurs by scoring each candidate cluster with a preference function that quantifies prior belief about the mixture composition. Clustering proceeds by selecting candidates to maximize their total preference score. SCAMP concludes by annotating each selected cluster with labels that describe how cluster-level statistics compare to certain dataset-level quantities. SCAMP can be run multiple times on a single data matrix. Comparison of annotations obtained across iterations provides a measure of clustering uncertainty. Simulation studies and applications to real data are considered. A C++ implementation with R interface is $\href{https://github.com/RGLab/scamp}{available\ online}$.