Aggregated data is commonplace in areas such as epidemiology and demography. For example, census data for a population is usually given as averages defined over time periods or spatial resolutions (city, region or countries). In this paper, we present a novel multi-task learning model based on Gaussian processes for joint learning of variables that have been aggregated at different input scales. Our model represents each task as the linear combination of the realizations of latent processes that are integrated at a different scale per task. We are then able to compute the cross-covariance between the different tasks either analytically or numerically. We also allow each task to have a potentially different likelihood model and provide a variational lower bound that can be optimised in a stochastic fashion making our model suitable for larger datasets. We show examples of the model in a synthetic example, a fertility dataset and an air pollution prediction application.

Existing graph-based methods for extractive document summarization represent sentences of a corpus as the nodes of a graph or a hypergraph in which edges depict relationships of lexical similarity between sentences. Such approaches fail to capture semantic similarities between sentences when they express a similar information but have few words in common and are thus lexically dissimilar. To overcome this issue, we propose to extract semantic similarities based on topical representations of sentences. Inspired by the Hierarchical Dirichlet Process, we propose a probabilistic topic model in order to infer topic distributions of sentences. As each topic defines a semantic connection among a group of sentences with a certain degree of membership for each sentence, we propose a fuzzy hypergraph model in which nodes are sentences and fuzzy hyperedges are topics. To produce an informative summary, we extract a set of sentences from the corpus by simultaneously maximizing their relevance to a user-defined query, their centrality in the fuzzy hypergraph and their coverage of topics present in the corpus. We formulate a polynomial time algorithm building on the theory of submodular functions to solve the associated optimization problem. A thorough comparative analysis with other graph-based summarization systems is included in the paper. Our obtained results show the superiority of our method in terms of content coverage of the summaries.

In 2016, the Johannesburg team at IBM Research discovered that the process of reporting cancer data to the government, which used it to inform national health policies, took four years after diagnosis in hospitals. In the US, the equivalent data collection and analysis takes only two years. The additional lag turned out to be due in part to the unstructured nature of the hospitals' pathology reports. Human experts were reading each case and classifying it into one of 42 different cancer types, but the free-form text on the reports made this very time-consuming. So the researchers went to work on a machine-learning model that could label the reports automatically.

While the fast-paced inception of novel tasks and new datasets helps foster active research in a community towards interesting directions, keeping track of the abundance of research activity in different areas on different datasets is likely to become increasingly difficult. The community could greatly benefit from an automatic system able to summarize scientific results, e.g., in the form of a leaderboard. In this paper we build two datasets and develop a framework (TDMS-IE) aimed at automatically extracting task, dataset, metric and score from NLP papers, towards the automatic construction of leaderboards. Experiments show that our model outperforms several baselines by a large margin. Our model is a first step towards automatic leaderboard construction, e.g., in the NLP domain.

TEHRAN - Iran's Revolutionary Guard shot down a U.S. surveillance drone Thursday in the Strait of Hormuz, marking the first time the Islamic Republic directly attacked the American military amid tensions over Tehran's unraveling nuclear deal with world powers. The two countries disputed the circumstances leading up to an Iranian surface-to-air missile bringing down the U.S. Navy RQ-4A Global Hawk, an unmanned aircraft with a wingspan larger than a Boeing 737 jetliner and costing over $100 million. Iran said the drone "violated" its territorial airspace, while the U.S. called the missile fire "an unprovoked attack" in international airspace over the narrow mouth of the Persian Gulf and President Donald Trump tweeted that "Iran made a very big mistake!" Trump later appeared to play down the incident, telling reporters in the Oval Office that he had a feeling that "a general or somebody" being "loose and stupid" made a mistake in shooting down the drone. The incident immediately heightened the crisis already gripping the wider region, which is rooted in Trump withdrawing the U.S. a year ago from Iran's 2015 nuclear deal and imposing crippling new sanctions on Tehran.

In this paper, we study how to learn an appropriate lane changing strategy for autonomous vehicles by using deep reinforcement learning. We show that the reward of the system should consider the overall traffic efficiency instead of the travel efficiency of an individual vehicle. In summary, cooperation leads to a more harmonic and efficient traffic system rather than competition

In recent years, art-creating AI has pushed the boundaries of how we define art. AI Expo Africa, the largest business focused AI event in Africa, have launched a grand challenge to create an original piece of artwork or music that leverages AI. There will be one winner for each category (visual art & music) and winners must be citizens of an African country and must be living in Africa. Awards will be presented to the winner at the exclusive AI Expo Africa VIP event on the opening night of AI Expo Africa in Cape Town, South Africa, on 3 September 2019 – with the winning music track(s) being played during the event and art work on display. Both works will be auctioned during an exclusive event on 4 September.

The MTN Group has launched Africa's first Mobile Money (MoMo) Artificial Intelligence (AI) service or "chatbot". A statement issued by the Group's Corporate Affairs on Tuesday said the chatbot went live in Ivory Coast in May and would be rolled out across MTN's MoMo footprint in the next few months. The AI mobile money "assistant" enables customers to engage with MTN's MoMo services, including payments on various social media platforms such as WhatsApp and Facebook Messenger, and via SMS. The statement said the service would also be included over time, in MTN's own newly released advanced instant messaging service "Ayoba". It said the chatbot was an AI guide that assists users to navigate MTN's MoMo services and provide other useful information.

Facebook is working closely with key non-profit and research partners to use artificial intelligence (AI) and big data to address large-scale social, health and infrastructure challenges in sub Saharan Africa. These efforts range from rural electrification in Tanzania to vaccinating people in remote corners of Malawi. Facebook is applying the processing muscle of its compute power, its extensive data science skills and its expertise in AI and machine learning to create the world's most detailed and accurate maps of local populations. Facebook also partners with Columbia University's Center for International Earth Science Information Network (CIESIN (http://www.ciesin.org/)) to ensure that this effort leverages the best available administrative data for all countries involved. The Boston-based Facebook team uses advanced computer vision and machine learning to combine satellite imagery from Digital Globe with public census data and other sources to create detailed population density maps of Africa.

Powered by AI technologies, businesses are transforming the way they operate. But what is the real opportunity for your business? Where can you implement tech to maximise the benefits and how do you need to adapt to take advantage? These are the questions being addressed at the cutting-edge of business transformation in Africa. Engaging with leading global multinationals as well as regional powerhouses, the AI Summit Cape Town event will uncover the true opportunity AI presents for the most forward-thinking enterprises, tackling the challenges around people-change management, upskilling the workforce, creating deeper relationships with customers and partners as well as understanding the business case.