Learning low-dimensional embeddings of knowledge graphs is a powerful approach used to predict unobserved or missing edges between entities.

President Joe Biden's surprise announcement Monday night--that a U.S. drone strike over the weekend killed Ayman al-Zawahiri, leader of al-Qaida and co-architect of the 9/11 terrorist attack--is both more and less significant than it might seem at first glance. On the one hand, mainly because of the West's counter-terrorism strategies, al-Qaida is far from the potent global force that it was a decade ago. Its presence has been muted, and Zawahiri himself has hidden so far out of sight that one prominent expert speculated back in November that he might have been killed already. On the other hand, one fact about this drone strike hints at a much larger finding: It took place in Afghanistan. It turns out Zawahiri was living with his family in a large safehouse in downtown Kabul--meaning he had to be there with the Taliban's full blessing. This means that, contrary to the Taliban's assurances, they have been plotting a revival of their alliance with al-Qaida--the alliance that Osama bin Laden formed at the turn of the century and that spawned the attack on the World Trade Center.

Attempts to render deep learning models interpretable, data-efficient, and robust have seen some success through hybridisation with rule-based systems, for example, in Neural Theorem Provers (NTPs). These neuro-symbolic models can induce interpretable rules and learn representations from data via back-propagation, while providing logical explanations for their predictions. However, they are restricted by their computational complexity, as they need to consider all possible proof paths for explaining a goal, thus rendering them unfit for large-scale applications. We present Conditional Theorem Provers (CTPs), an extension to NTPs that learns an optimal rule selection strategy via gradient-based optimisation. We show that CTPs are scalable and yield state-of-the-art results on the CLUTRR dataset, which tests systematic generalisation of neural models by learning to reason over smaller graphs and evaluating on larger ones. Finally, CTPs show better link prediction results on standard benchmarks in comparison with other neural-symbolic models, while being explainable. All source code and datasets are available online, at https://github.com/uclnlp/ctp.

Neuro-symbolic and statistical relational artificial intelligence both integrate frameworks for learning with logical reasoning. This survey identifies several parallels across seven different dimensions between these two fields. These cannot only be used to characterize and position neuro-symbolic artificial intelligence approaches but also to identify a number of directions for further research.

Interview Natural language processing, or NLP for short, is one of the key research areas of artificial intelligence, and has had a major boost in the last couple of years. After all, the written word is everywhere – be it in the form of scientific texts, news articles, wiki entries, or seemingly simply status updates on the social platform of your choice – and there's a huge interest in making sense of it all. One company especially enthusiastic about NLP is Facebook. No matter your opinion about its platform, the amount of data the Silicon Valley giant has to process every single day is simply mind boggling. And its platform comes with some NLP already built in: take its translation capabilities, for example.

Latent factor models are increasingly popular for modeling multi-relational knowledge graphs. By their vectorial nature, it is not only hard to interpret why this class of models works so well, but also to understand where they fail and how they might be improved. We conduct an experimental survey of state-of-the-art models, not towards a purely comparative end, but as a means to get insight about their inductive abilities. To assess the strengths and weaknesses of each model, we create simple tasks that exhibit first, atomic properties of binary relations, and then, common inter-relational inference through synthetic genealogies. Based on these experimental results, we propose new research directions to improve on existing models.

Many problems on signal processing reduce to nonparametric function estimation. We propose a new methodology, piecewise convex fitting (PCF), and give a two-stage adaptive estimate. In the first stage, the number and location of the change points is estimated using strong smoothing. In the second stage, a constrained smoothing spline fit is performed with the smoothing level chosen to minimize the MSE. The imposed constraint is that a single change point occurs in a region about each empirical change point of the first-stage estimate. This constraint is equivalent to requiring that the third derivative of the second-stage estimate has a single sign in a small neighborhood about each first-stage change point. We sketch how PCF may be applied to signal recovery, instantaneous frequency estimation, surface reconstruction, image segmentation, spectral estimation and multivariate adaptive regression.

However, technologies that mimic social cues and dialogue can already be seen today, Riedel said. "What the AI looks like in'Westworld' is in some ways what game companies are trying to do in virtual worlds," Riedel said. In the show "Westworld," the hosts are given pre-installed storylines that are triggered when a person interacts with them. It's similar to how game mechanics work in action role-playing video games like "Borderlands," Riedel said. Still, "Westworld's" missions are much more complex than what our current technology enables, he said.

Relation extraction, which learns semantic relations of concept pairs from text, is an approach for mining commonsense knowledge. This paper investigates an approach for relation extraction, which helps expand a commonsense knowledge base with little labor work. We proposed a framework that learns new pairs from Chinese corpora by adopting concept pairs in Chinese commonsense knowledge base as seeds. Multiple instance learning is utilized as the learning algorithm for predicting relation for unseen pairs. The performance of our system could be improved by learning multiple iterations. The results in each iteration are manually evaluated and processed to next iteration as seeds. Our experiments extracted new pairs for relations “AtLocation”, “CapableOf”, and “HasProperty”. This study showed that new pairs could be extracted from text without huge humans work.

We speed up marginal inference by ignoring factors that do not significantly contribute to overall accuracy. In order to pick a suitable subset of factors to ignore, we propose three schemes: minimizing the number of model factors under a bound on the KL divergence between pruned and full models; minimizing the KL divergence under a bound on factor count; and minimizing the weighted sum of KL divergence and factor count. All three problems are solved using an approximation of the KL divergence than can be calculated in terms of marginals computed on a simple seed graph. Applied to synthetic image denoising and to three different types of NLP parsing models, this technique performs marginal inference up to 11 times faster than loopy BP, with graph sizes reduced up to 98%--at comparable error in marginals and parsing accuracy. We also show that minimizing the weighted sum of divergence and size is substantially faster than minimizing either of the other objectives based on the approximation to divergence presented here.