Knowledge bases (KBs) are the backbone of many ubiquitous applications and are thus required to exhibit high precision. However, for KBs that store subjective attributes of entities, e.g., whether a movie is "kid friendly", simply estimating precision is complicated by the inherent ambiguity in measuring subjective phenomena. In this work, we develop a method for constructing KBs with tunable precision--i.e., KBs that can be made to operate at a specific false positive rate, despite storing both difficult-to-evaluate subjective attributes and more traditional factual attributes. The key to our approach is probabilistically modeling user consensus with respect to each entity-attribute pair, rather than modeling each pair as either True or False. Uncertainty in the model is explicitly represented and used to control the KB's precision. We propose three neural networks for fitting the consensus model and evaluate each one on data from Google Maps--a large KB of locations and their subjective and factual attributes. The results demonstrate that our learned models are well-calibrated and thus can successfully be used to control the KB's precision. Moreover, when constrained to maintain 95% precision, the best consensus model matches the F-score of a baseline that models each entity-attribute pair as a binary variable and does not support tunable precision. When unconstrained, our model dominates the same baseline by 12% F-score. Finally, we perform an empirical analysis of attribute-attribute correlations and show that leveraging them effectively contributes to reduced uncertainty and better performance in attribute prediction.

This paper focuses on the end-to-end abstractive summarization of a single product review without supervision. We assume that a review can be described as a discourse tree, in which the summary is the root, and the child sentences explain their parent in detail. By recursively estimating a parent from its children, our model learns the latent discourse tree without an external parser and generates a concise summary. We also introduce an architecture that ranks the importance of each sentence on the tree to support summary generation focusing on the main review point. The experimental results demonstrate that our model is competitive with or outperforms other unsupervised approaches. In particular, for relatively long reviews, it achieves a competitive or better performance than supervised models. The induced tree shows that the child sentences provide additional information about their parent, and the generated summary abstracts the entire review.

Knowledge graphs capture structured information and relations between a set of entities or items. As such knowledge graphs represent an attractive source of information that could help improve recommender systems. However, existing approaches in this domain rely on manual feature engineering and do not allow for an end-to-end training. Here we propose Knowledge-aware Graph Neural Networks with Label Smoothness regularization (KGNN-LS) to provide better recommendations. Conceptually, our approach computes user-specific item embeddings by first applying a trainable function that identifies important knowledge graph relationships for a given user. This way we transform the knowledge graph into a user-specific weighted graph and then apply a graph neural network to compute personalized item embeddings. To provide better inductive bias, we rely on label smoothness assumption, which posits that adjacent items in the knowledge graph are likely to have similar user relevance labels/scores. Label smoothness provides regularization over the edge weights and we prove that it is equivalent to a label propagation scheme on a graph. We also develop an efficient implementation that shows strong scalability with respect to the knowledge graph size. Experiments on four datasets show that our method outperforms state of the art baselines. KGNN-LS also achieves strong performance in cold-start scenarios where user-item interactions are sparse.

Today is the slowest business will ever run. In today's world, it is the small companies that beat the large enterprises with speed and agility, according to Valerie Blatt – Global Vice President SAP. Companies such as Spotify are forcing incumbents such as Apple to change their music business. Therefore, for the organisation of tomorrow to remain relevant, they need faster, better and more intelligence. Last week, I was invited to join SAP Ariba Live in Barcelona.

How to ensure human autonomy over our computational "screens, scenes, and unseens." Digital assistants such as Alexa and Siri and Google Assistant can be quite helpful -- but their actual allegiance is to Amazon and Apple and Google, not to the ordinary people who use them. By introducing AI-based digital agents that truly represent and advocate for us as individuals, rather than corporate or government institutions, we can make the Web a more trustworthy and accountable place. In the 2004 film "I Robot," Will Smith's character, the enigmatic Detective Del Spooner, harbors an animosity toward the humanoid-like robots operating in his society. Over the course of the film we learn why.

The Nordic nation has pledged to go carbon neutral by 2035 – not bad for one of the coldest countries in the world. The Hollywood star says he wants to solve climate change with robots. Details are scant, but come on, he's Iron Man! Millions of people are rejoicing as iTunes shuffles of this mortal coil. Apple says the long-hated app will still survive on Windows, however. Forget "my kid could draw that" – now robots are producing bad art.

Get some huge savings on all your favorite electronics, like a new Kindle or robot vacuum. If you make a purchase by clicking one of our links, we may earn a small share of the revenue. However, our picks and opinions are independent from USA Today's newsroom and any business incentives. You know that scene in 30 Rock when Liz Lemon is like "What a week, huh?" and Alec Baldwin's character goes, "Lemon, it's Wednesday." I happen to know a great cure to the Hump Day blues however, and it involves doing a bit of retail therapy. I don't just recommend buying a bunch of stuff you don't need at Target or on Amazon.

Deep fakes are "videos that have been constructed to make a person appear to say or do something that they never said or did. Many commentators worry that voters will be influenced by pure fiction: As the 2020 US election looms, we learn of the fight to stay ahead (CNN, April 26, 2019) of the "growing threat" (ABC Eyewitness News, May 7, 2019) because "The 2020 campaigns aren't ready for deepfakes" (Axios/HBO, June 4, 2019). We Can No Longer Believe What We See, the New York Times warned yesterday. Under the right set of circumstances, deepfakes will be very influential. They don't even have to be particularly good to potentially swing the outcome of an election. As with so much in elections, deepfakes are a numbers game. While the presence of tampering in all but the most sophisticated deepfakes can be quickly identified, not everyone who views them will get that message. More fundamentally, not everyone wants to get that message. As can occur with other forms of online ...

AI is rapidly changing everything. It is transforming society, the very way we live and act as social creatures, how we behave, how we do business, and even the very fabric of our own human identity. Cities are managed by machine driven big data gathered by sensors, constructing together what we known as the Internet of Things. An escalating digital transformation is transforming our vast amounts of paper ledgers into digital records, from traffic to finance to medical records. These, which convey most of the data gathered about us, are now processed in the cloud and augmented with machine learning artificial intelligence's multiple tools, which are stored in blockchain distributed ledgers.

This thesis presents new methods for unsupervised learning of distributed representations of words and entities from text and knowledge bases. The first algorithm presented in the thesis is a multi-view algorithm for learning representations of words called Multiview Latent Semantic Analysis (MVLSA). By incorporating up to 46 different types of co-occurrence statistics for the same vocabulary of english words, I show that MVLSA outperforms other state-of-the-art word embedding models. Next, I focus on learning entity representations for search and recommendation and present the second method of this thesis, Neural Variational Set Expansion (NVSE). NVSE is also an unsupervised learning method, but it is based on the Variational Autoencoder framework. Evaluations with human annotators show that NVSE can facilitate better search and recommendation of information gathered from noisy, automatic annotation of unstructured natural language corpora. Finally, I move from unstructured data and focus on structured knowledge graphs. I present novel approaches for learning embeddings of vertices and edges in a knowledge graph that obey logical constraints.